T/I B RAR.Y OF THE UNIVERSITY OF ILLINOIS 590.5 FI v. 35 cop. 3 HATURAL HISfORY SURVEY I 3 BIRDS OF NORTHEASTERN PERU MELVIN A. TRAYLOR 1 n .,.,-, FIELDIANA: ZOOLOGY VOLUME 35, NUMBER 6 Published by CHICAGO NATURAL HISTORY MUSEUM JUNE 30, 1958 •w HISTORY SURVEY BIRDS OF NORTHEASTERN PERU MELVIN A. TRAYLOR Assistant Curator, Division of Birds FIELDIANA: ZOOLOGY VOLUME 35, NUMBER 5 Published by CHICAGO NATURAL HISTORY MUSEUM JUNE 30, 1958 Library of Congress Catalog Card Number: 5 8-1 SI 04 PRINTED IN THE UNITED STATES OF AMERICA BY CHICAGO NATURAL HISTORY MUSEUM PRESS Birds of Northeastern Peru1 The present report is an annotated list of birds collected by Jos4 M. Schunke in northern Peru, from 1945 to 1947. The majority of the specimens are from localities along the Ucayali River in the Department of Loreto; the remainder are from the divide between the Ucayali and Huallaga Rivers in the Department of Huanuco. The localities along the Ucayali, all of which are well within the humid tropical zone, are: Contamana: a town on the right bank of the Ucayali, at about 7° 16' S. Lat. Cerro Azul: about fifteen miles east of Contamana at 500 meters elevation; this is not the Cerro Azul of the millionth map (American Geographic Society), a locality which is northwest of Contamana and much higher. Pucallpa: a town on the left bank of the river at about 8° 25' S. Lat. Yarinacocha: a small laguna about fifteen miles north by west of Pucallpa. Rio Pachutia: probably the Rio Pachitea, a river flowing into the Ucayali about thirty miles south of Pucallpa. Previsto and Quistacocha: unidentified localities which must be near Pucallpa, to judge from the collecting dates. Only one specimen was collected at each of the last three localities. The two localities on the divide between the Ucayali and the Hua- llaga— Divisoria and Fundo Sinchona — for practical purposes may be considered the same. They are on the crest of the ridge where the road from Pucallpa to Tingo Maria, a small town on the Hua- llaga River, crosses the divide. The altitude at the crest is about 1,600 meters, and Schunke collected from about 1,300 meters to the top. These localities are in the subtropical zone, and only a very few of the tropical forms from the Ucayali were found. The collection as a whole numbers 807 specimens, of which 689 are from the Ucayali and 118 from the divide. The total species from the Ucayali are 236 and from the divide 69. The total number of forms, however, is 301, which leaves only four forms common to both zones. Two of these are hawks, Harpagus bidentatus and Buteo magnirostris, the third is the wide-ranging squirrel cuckoo, Piaya cayana, and the fourth is Cercomacra nigrescens. This almost com- 1 Manuscript submitted October, 1956. 87 88 FIELDIANA: ZOOLOGY, VOLUME 35 plete separation between the species in these two collections from humid forests less than a hundred miles apart further demonstrates the reality of the Andean life zones. The only previous report on the birds of northeastern Peru was that published by Sclater and Salvin (1873), which summarized the collections of Bartlett, Hauxwell and Bates. The recent papers on Peruvian birds by Zimmer have added many new forms to the list of species known from the region, and have been of great aid in the identification of forms in the families which he has covered. Gylden- stolpe's recent paper on the birds of the Rio Jurua of western Brazil (1945) has been particularly valuable since in most cases the same forms are found on the Ucayali and the upper Jurua. During the preparation of this report comparative material was borrowed from the American Museum of Natural History and from the United States National Museum, and I wish to thank the au- thorities of these two institutions for their generous help. I would also like to thank Dr. Austin L. Rand and Mr. Emmet R. Blake of Chicago Natural History Museum for their constant help through- out the preparation of this paper. All measurements are in millimeters. The wings were measured flat, and the culmen was measured from the base unless otherwise stated. Systematic List Family TINAMIDAE Tinamus tao kleei Tschudi Cerro Azul, 1 9 . This specimen is almost as gray above as typical tao from the Rio Tapajoz, but below it has the brownish wash characteristic of kleei. Bond and de Schauensee (1943, p. 168) concluded that wed- delli Bonaparte of Bolivia is separable from kleei on the basis of greater size. They based their conclusions on a comparison of a male and female from Palmar, Cochabamba, with a male and female from La Pampa, Sandia, in southern Peru. The comparative meas- urements were: Exp. Species Sex Wing Culmen Tarsus weddelli (Palmar) d" 270 36 86 9 268 36 87 kleei (Sandia) cf 244 31.5 69 9 230 30 64 TRAYLOR: BIRDS OF NORTHEASTERN PERU 89 Specimens in the Conover collection, however, fail to show these differences. Two males from Bolivia are in substantial agreement with Bond and de Schauensee's birds: wing 291, 269; exposed cul- men 35, 36; tarsus 85, 74. A female and an unsexed specimen from Cerro Azul and the Rio Urubamba, Peru, are closely similar: wing 263, 288; exposed culmen 37, 37; tarsus 79, 81. A male and a female from Cututcu, eastern Ecuador, also agree: wing 271, 290; exposed culmen 34, 33.5; tarsus 81, 88. From Bolivia to eastern Ecuador, therefore, there is no significant difference in size or color, and this whole population should bear the name kleei, with weddelli as a syn- onym. The much smaller size of the birds from Sandia is due to immaturity, according to Conover, who has examined them critically. In a more recent paper (1955b, p. 207) Bond has reached the same conclusion, and he makes weddelli a synonym of kleei. Tinamus major ruficeps Sclater and Salvin Yarinacocha, 1 d" ; Pucallpa, 2 downy 9 . Blake (1955, p. 10) and Bond (1955b, p. 208) have both come to the conclusion that peruvianus cannot be maintained as a valid race. Tinamus guttatus Pelzeln Yarinacocha, 1 cf , 2 9. Crypturellus cinereus cinereus Gmelin Yarinacocha, 2 d", 3 9 . Crypturellus soui inconspicuus Carriker Yarinacocha, 1 c? , 3 9 , 1 d" imm., 1 9 juv. One female shows rufescent coloration approaching that of nigri- ceps of eastern Ecuador, but the other adult females and the adult male are similar to a series of inconspicuus from Bolivia. Crypturellus undulatus yapura Spix Yarinacocha, 1 9 . Crypturellus variegatus salvini Salvador! Cerro Azul, 1 9 . This specimen agrees with a series of four birds from eastern Ecuador. 90 FIELDIANA: ZOOLOGY, VOLUME 35 Crypturellus brevirostris bartletti Sclater and Salvin Yarinacocha, 2 cf , 1 9,2 9 imm. Family ANHINGIDAE Anhinga anhinga anhinga Linnaeus Yarinacocha, 1 cf , 1 9 imm. Family ARDEIDAE Butorides striatus striatus Linnaeus Yarinacocha, 2 9 . Nycticorax nycticorax hoactli Gmelin Yarinacocha, 4 cf , all imm. Tigrisoma lineatum lineatum Boddaert Yarinacocha, 1 cf ; Contamana, 1 cf imm. Family ANHIMIDAE Anhima cornuta Linnaeus Yarinacocha, 2 cf , 1 9. Family ACCIPITRIDAE Elanoides forficata yetapa Vieillot Yarinacocha, 1 9 . Leptodon cayanensis Latham Yarinacocha, 1 cf . Harpagus bidentatus bidentatus Latham Yarinacocha, 1 cf , 2 9 ; Pucallpa, 1 9 ; Fundo Sinchona, 1 cf . Birds from upper Amazonia (eastern Colombia, Ecuador, Peru, Bolivia, and the upper Rio Jurua, Brazil) are larger than those from eastern South America. The wing measurements for adults and im- matures are given separately below, since the immature birds average smaller than adults. TRAYLOR: BIRDS OF NORTHEASTERN PERU 91 Adults 9 9 Upper Amazonia .................. (6) 214-222 (217.5) (4) 218-232 (224) Eastern South America ............ (1) 205 (3) 204-213 (210) Immatures Upper Amazonia .................. (2)192-203(200) (1)213 Eastern South America ............ (2) 199-215 (207) (7) 193-215 (204) Although these differences seem consistent except among the im- matures, the sample is too small to provide a basis for a new race. Accipiter bicolor bicolor Vieillot Yarinacocha, 1 9 . Accipiter superciliosus superciliosus Linnaeus Yarinacocha, 1 d", 1 9 ; Cerro Azul, 1 d". This is apparently a rare bird in Peru. The only previous record was of a female from Moyabamba and a male from Rioja, both speci- mens in Chicago Natural History Museum and both listed by Hell- mayr and Conover (1949, p. 64). Buteo albonotatus Kaup Yarinacocha, 3 d", 1 9 . Hellmayr and Conover (1949, p. 154) do not recognize the race abbreviatus of South America, pointing out that many "abbreviatus" were probably winter migrants from North America, and that the species is not known certainly to breed east of the Andes. The three males from Yarinacocha were collected in December and Jan- uary and could well be migrants; the female was collected Septem- ber 17, which would be early. South American specimens do average smaller in wing length. Combining Hellmayr and Conover's (1949, p. 157) measurements with my own, I have: Wing North and Central America ................... 12 cf d" 384-411 (av. 395) 699 410-431 South America .............................. 7 d" d" 367-386 (av. 381) 299 398, 460(!) Despite the exceptional Bolivian female with a wing of 460, the South American birds show a consistent enough reduction in wing length to at least keep open the question of the validity of abbrevi- atus. 92 FIELDIANA: ZOOLOGY, VOLUME 35 Buteo magnirostris occiduus Bangs Yarinacocha, 3 d", 3 9 ; Fundo Sinchona, 1 d*. Busarellus nigricollis nigricollis Latham Yarinacocha, 2 9 . Spizastur melanoleucus Vieillot Yarinacocha, 1 d* , 1 9. This is the first record of this species for Peru, and a considerable extension of range for the form. Previously, its known range in South America included the Guianas, eastern Brazil, Paraguay and northern Argentina. Spizaetus ornatus ornatus Daudin Yarinacocha, 1 cf. Geranospiza caerulescens caerulescens Vieillot Contamana, 1 9 . This is apparently the first record of this race from Peru. The Geranospiza caerulescens of Taczanowski (1884, p. 168) from Lech- ugal in extreme northwestern Peru belongs to the race balzarensis W. L. Sclater of western Ecuador. The latter is a much larger bird, the wing of the female, from Taczanowski, being 322, while in the present specimen of caerulescens it is only 268. Family PANDIONIDAE Pandion haliaetus carolinensis Gmelin Yarinacocha, 2 d". Family FALGONIDAE Herpetotheres cachinnans cachinnans Linnaeus Yarinacocha, 4 cf , 1 9 . Although Peru has been considered to be within the range of the pale southern race queribundus, these specimens are as dark as a series of cachinnans from the Guianas. A comparison of birds from Paraguay and Bolivia with those from Venezuela and the Guianas indicates that queribundus is at best a weakly distinguished race. Hellmayr and Conover have synonymized it with cachinnans (1949, p. 237). TRAYLOR: BIRDS OF NORTHEASTERN PERU 93 Micrastur semitorquatus semitorquatus Vieillot Yarinacocha, 2 cf . Micrastur buckleyi Swarm Yarinacocha, 1 d". In a recent publication (1948, p. 199) I have discussed this form more fully and shown why I consider it a full species. Its resem- blance to semitorquatus in color is remarkable, but the differences in size and proportions can only be accounted for on the basis of two separate species. Micrastur gilvicollis gilvicollis Vieillot Contamana, 1 d". Daptrius ater Vieillot Yarinacocha, 2 cf, 1 9 imm. Family CRACIDAE Mitu mitu Linnaeus Yarinacocha, 1 9 . Chamaepetes goudotii tschudii Taczanowski Fundo Sinchona, 1 9 . This specimen is definitely tschudii of northern Peru and Ecuador and not rufiventris of central Peru. The latter has been taken at Cushi Libertad in southern Huanuco, not far from Fundo Sinchona but on the opposite side of the Huallaga. Pipile cumanensis cumanensis Jacquin Yarinacocha, 1 9 . Aburria aburri Lesson Fundo Sinchona, 2 d", 1 9 ; Divisoria, 1 9 . Family PHASIANIDAE Odontophorus speciosus speciosus Tschudi Fundo Sinchona, 1 d". 94 FIELDIANA: ZOOLOGY, VOLUME 35 The males of the nominate race are most easily separated from soderstromii of eastern Ecuador by the heavy white streaking above, and the dark auriculars. Hellmayr and Conover (1942, p. 275) state that the female of speciosus is unknown. There are at present two topotypical females of this race from Chanchamayo in the Conover collection. Above, they are like the males except that they lack the white shaft streak- ing. Below, the throat is black, the black extending up the sides of the head and including the auriculars, and the upper breast is bright chestnut, this color extending up the sides of the neck. The rest of the under parts are slate gray, many of the feathers being lightly edged with chestnut. The females of soderstromii (three specimens from eastern Ecua- dor) are best separated from females of speciosus by the darker, more reddish brown of the upper parts and by the reddish auriculars. Hellmayr and Conover stated that the throat was black in the females, but the present specimens show a range of variation sim- ilar to that of the males, from pure black to auburn with only a few black bars on the lower throat. The rest of the under parts likewise vary from a clear brownish gray to a lighter gray tinged with chestnut. Odontophorus stellatus Gould Yarinacocha, 5 d", 1 9, 1 d* imm. The gray on the hind neck and upper back extends considerably farther down the back in Peruvian specimens than in birds from Ecuador and Brazil. Family PSOPHIIDAE Psophia leucoptera leucoptera Spix Cerro Azul, 2 d", 1 9. Family RALLIDAE Aramides cajanea cajanea Miiller Yarinacocha, 2 9 . Laterallus fasciatus Sclater and Salvin Yarinacocha, 1 cf, 1 9 ; Pucallpa, 1 cf, 1 9 TRAYLOR: BIRDS OF NORTHEASTERN PERU 95 Porphyrula martinica Linnaeus Yarinacocha, 1 9 imm. Family HELIORNITHIDAE Heliornis fulica Boddaert Yarinacocha, 1 d" , 1 9 . Family EURYPYGIDAE Eurypyga helias helias Pallas Yarinacocha, 1 cf. Family JACANIDAE Jacana spinosa peruviana Zimmer Yarinacocha, 1 d" , 1 9, 2 o" imm., 1 9 imm. The two adult birds agree well with the type of peruviana, dif- fering only in being slightly darker below. They are much darker than the neighboring race jacana but are poorly differentiated from the more distant race intermedia of Colombia and Venezuela (cf. Hellmayr and Conover, 1948, p. 9). Family SCOLOPACIDAE Tringa solitaria solitaria Wilson Pucallpa, 1 cf . Tringa solitaria cinnamomea Brewster Pucallpa, 1 9 imm. Family RECURVIROSTRIDAE Himantopus himantopus melanurus Vieillot Yarinacocha, 1 cf . Although the white band across the back is only partially formed, the large amount of white on the head places this specimen nearer to melanurus than to mexicanus of North America and northern South America. The Ucayali valley seems to be the meeting ground 96 FIELDIANA: ZOOLOGY, VOLUME 35 for these two races, since Zimmer (1930, p. 255) had an undoubted specimen of mexicanus from Puerto Bermudez. Family LARIDAE Phaetusa simplex Gmelin Yarinacocha, 1 cf , 2 9 . See Hellmayr and Conover (1948, p. 296, footnote) for the use of the binomial. Family COLUMBIDAE Columba cayennensis sylvestris Vieillot Yarinacocha, 3 d" , 2 9 . Columba subvinacea recondita Todd Yarinacocha, 2 d", 2 9 . Although the whole of Peru has previously been considered with- in the range of ogilvie-granti, type locality Guayabamba, Peru, the present series is inseparable from a series of recondita from the Rio Tapajoz. This is also true of a pair from the lowlands of eastern Ecuador. All previous records of ogilvie-granti from Ecuador and Peru are from above 3,000 feet, and it appears to be a race of the lower subtropical zone, whose place in the tropics is taken by recon- dita of Amazonian Brazil. In 1945 Gyldenstolpe (1945, p. 46) tenta- tively assigned specimens from the upper Jurua to ogilvie-granti, but more recently (1951, p. 58) he has reconsidered and placed them in recondita. In Bolivia ogilvie-granti descends to the tropics in the departments of Beni and Santa Cruz. Columba plumbea subsp. Yarinacocha, 2 cf . One of these specimens is typical of pallescens of Amazonia, and the other agrees with bogotensis of the eastern Andes. The former has been reported from Puerto Indiana on the lower Ucayali and the latter from Chanchamayo on the upper Ucayali drainage, so these specimens are intermediate geographically and they also indicate a mixed population morphologically. Columba plumbea bogotensis Berlepsch and Leverkiihn Fundo Sinchona, 1 d". TRAYLOR: BIRDS OF NORTHEASTERN PERU 97 This specimen agrees well with birds from the Chanchamayo valley; the separation between bogotensis and pallescens may prove to be altitudinal in eastern Peru. Columbigallina talpacoti talpacoti Temminck Yarinacocha, 2 cf , 2 9 . Claravis pretiosa Ferrari-Perez Yarinacocha, 1 cf , 1 9 . Leptotila verreauxi decipiens Salvadori Yarinacocha, 1 cf , 1 9. Leptotila rufaxilla dubusi Bonaparte Yarinacocha, 1 9 . Osculatia saphirina rothschildi Sztolcman Cerro Azul, 1 cf . The race rothschildi was separated from the nominate race of eastern Ecuador on the basis of the following characters, shown by a single female from Cadena, Marcapata Valley, Cuzco: lower back and rump more bluish, less purplish; white spot on secondaries larger; size smaller. The present specimen and a male from the Marcapata Valley amply confirm the validity of this race. The most constant character is the color of the back, which is much more bluish-, less reddish-, purple in rothschildi than in saphi- rina. The extent of the white spot on the secondaries is more vari- able. In saphirina the spot is confined to one feather and measures about 2x4 mm. In the topotype of rothschildi the spot is similar to that in saphirina, but the Cerro Azul male has a white spot about 5x6 mm. on the innermost secondary, a 2x4 mm. spot on the ad- joining feather, and white edgings on two of the others. In size, rothschildi seems to average smaller; wing measurements of 2 males are 134, 140; of five saphirina males from eastern Ecuador 138, 140, 141, 143, and 146. The Cerro Azul specimen gives a large extension of range for this form. Further intensive collecting in the tropical zone along the foot of the Andes will be needed to determine if the range is really as fragmented as it appears to be. 98 FIELDI ANA: ZOOLOGY, VOLUME 35 Geotrygon montana montana Linnaeus Yarinacocha, 2 cf, 1 9,1 9 imm. Family PSITTACIDAE Ara militaris militaris Linnaeus Fundo Sinchona, 1 9 . Ara macao Linnaeus Contamana, 1 9 . Ara manilata Boddaert Yarinacocha, 1 9 . Ara couloni Sclater Fundo Sinchona, 2 d". This species is apparently a representative of A. maracana, but the differences are too striking to be considered only of subspecific value. Aratinga weddelli Deville Yarinacocha, 3 d"1, 2 9 . Forpus xanthopterygius crassirostris Taczanowski Yarinacocha, 1 cT imm., 3 9 . Gyldenstolpe (1945, p. 52) shows that the races of the "passeri- nus" group of parrots which have bright blue rumps should be con- sidered a separate species to which the name xanthopterygius should be applied. Brotogeris versicolurus versicolurus Miiller Yarinacocha, 1 cf, 2 9 . Brotogeris gustavi cyanoptera Salvadori Yarinacocha, 3 d", 7 9 . Brotogeris g. cyanoptera is found throughout western Amazonia from southern Colombia through eastern Ecuador and Peru and western Brazil to northern and eastern Bolivia. Brotogeris g. gus- tavi, which is characterized by yellow on the bend of the wing and TRAYLOR: BIRDS OF NORTHEASTERN PERU 99 the carpal edge, has a very restricted range in the upper Huallaga valley in northern Peru. Hitherto the two forms have been consid- ered separate species. However, within a series of birds from Moyo- bamba and Yurimaguas on the lower Huallaga, two specimens have the bend of the wing and the carpal edge yellow, four show varying amounts of yellow on the wings, and the rest have plain green wings. Thus, the gap between gustavi and cyanoptera is fully bridged and the two forms must be considered conspecific, the specific name gustavi having priority. When the relationship between cyanoptera, beniensis, and chrysosema in northern Bolivia and western Matto Grosso is more fully explored it may be found that all three forms are conspecific. Pionites melanocephala pallida Berlepsch Yarinacocha, 1 d" , 3 9 . Compared with three males of P. m. melanocephala from British Guiana, these birds are strikingly different, being paler buff on the breast and abdomen and having the throat, lower flanks, and thighs a pure clear yellow. In melanocephala the feathers of the latter parts are a light orange brick color with narrow yellow tips, but in the Peruvian specimens they are pure yellow down to the base. A pair of birds from Sarayacu, Ecuador, however, is exactly intermediate between melanocephala and Peruvian pallida. They are darker buff below as in melanocephala, but have the flanks and thighs yellow with only a faint tint of reddish. Close examination shows, how- ever, that all the feathers are pinkish at the base, as opposed to the pure yellow feathers of the Peruvian birds, so that the difference in color is not due to varying degrees of wear. If larger series from Ecuador and Peru should show that these differences are constant, a third race would have to be separated. At present the type local- ity of pallida is "eastern Peru and eastern Ecuador." To avoid confusion in case it becomes necessary to describe a third race, I suggest that the type locality of pallida be restricted to Yurimaguas, Dept. San Martin, Peru, the only locality from which Berlepsch listed a specimen (1889, p. 317). Amazona festiva festiva Linnaeus Yarinacocha, 1 9 imm. Amazona amazonica amazonica Linnaeus Yarinacocha, 1 d" , 1 9 . 100 FIELDIANA: ZOOLOGY, VOLUME 35 These specimens differ from birds from the rest of South America in having a pale ashy blue wash to the plumage and in having the black edges on the feathers of the hind neck more distinct. The bills are also much larger: culmen from cere, d* 35, 9 34, as opposed to 9 d" cT 30-32, 899 27-31, for birds from Venezuela, the Guianas and Brazil. However, of the birds available for comparison only one female is from Amazonia, the restricted type locality of amazonica, so that it is not possible to say what separation, if any, is warranted. Griscom and Greenway's race, micro, (1937, p. 420) appears valid, but with its range restricted to Dutch Guiana and possibly eastward. A single male from Dutch Guiana has a wing of 198, within the range of micro, as defined by the authors, but two males from British Guiana have wings 212 and 219, well within the range of typical amazonica. Amazon a farinosa inornata Salvadori Yarinacocha, 1 cf; Cerro Azul, 1 cf, 1 9 . Wing lengths of 240-245 place these specimens in the subspecies inornata, rather than in the larger race chapmani. At the time of the original description of chapmani (Traylor, 1948, p. 195) I believed that its range was divided into two widely separated parts, eastern Ecuador and Bolivia respectively, with inornata being found in Brazil and Peru. Recently, however, Bond (1955b, p. 232) has re- corded a specimen of chapmani from El Tingo, on the Huallaga River, Dept. San Martin, and suggests that populations of this form will be found in Peru west of the eastern Andes. Family CUCULIDAE Coccyzus erythrophthalmus Wilson Yarinacocha, 1 cf. Coccyzus americanus subsp. Yarinacocha, 1 9 . The wing of this specimen is 147, halfway between the averages of americanus and occidentalis. Piaya cayana nigricrissa Cabanis Yarinacocha, 2 d", 2 9 ; Pucallpa, 1 9 ; Fundo Sinchona, 1 o"; Divisoria, 1 9 . TRAYLOR: BIRDS OF NORTHEASTERN PERU 101 Piaya minuta minuta Vieillot Yarinacocha, 1 d" juv. The juvenile plumage has been well described by Zimmer (1930, p. 261). Crotophaga major Gmelin Yarinacocha, 2 cf , 2 9 . Crotophaga ani Linnaeus Yarinacocha, 4 cf , 2 9 . Tapera naevia naevia Linnaeus Yarinacocha, 2 d", 1 9 . Since Bangs and Penard (1918, p. 50) resurrected the race chochi for the birds of southern South America on rather meager material there has been very little agreement among authors as to the validity of the race, or the range it supposedly occupies. Peters (1940, p. 58) gave its range as from Matto Grosso to northern Argentina. Bond, however (1955b, p. 235), includes birds from western Peru, western Ecuador, and Colombia within the race. He also questions whether excelkns of Mexico and Central America can be separated from chochi. The characters separating chochi from naevia of the Guianas, as given by Bangs and Penard, were greater size and more brownish coloration above. With 31 adult males available, covering the whole range of the species, it is evident that these differences are apparent on the average, but not to a degree sufficient to maintain chochi as a distinct subspecies. Specimens from the Guianas and Venezuela have the edges of the back feathers and the long upper tail coverts a pale grayish brown, without much warmth. They also average the smallest of any popu- lation (wing lengths of 8 cf