AMERICAN MUSEUM NOVITATES Published by Number 1168 THE AMERICAN MUSEUM OF NATURAL HISTORY April 21, 1942 New York City

STUDIES OF PERUVIAV BIRDS. NO. XLII' THE GENUS POLIOPTILA BY JOHN T. ZIMMER As in other papers of this series, names of I believe, therefore, that superciliaris is colors are capitalized when direct compari- a recognizable form, ranging from the Canal son has been made with Ridgway's "Color Zone to Guatemala. I am unable to Standards and Color Nomenclature." separate "magna" of Ridgway from Costa Rica, which I find to fall within the limits Polioptila plumbea bilineata (Bonaparte) of variation of superciliaris. (Culicivora) bilineata BONAPARTE (ex LICHTEN- In this connection, I find myself unable sTEIN MS.), 1860, Consp. Av., I, p. 316-Carth- to agree with Hellmayr who placed albiloris agena, Colombia; cotypes in Berlin Mus. and "bairdi" in the same species with A long series of birds from various parts bilineata and commented on certain ex- of Central America is sufficiently distinct amples which he considered intermediate from South American specimens to raise between the two groups. The distinctions the question, once again, of the validity between the two groups are no greater than of superciliaris and magna as separable those which separate certain of the con- forms. The birds from Central America species of bilineata from each other, and have the lateral under parts quite notice- the specific association of albiloris with the ably darker and grayer than the South same forms would be strongly suggested American examples of bilineata, and the except for an apparent conflict of ranges gray of the back averages a little darker, that appears in the material at hand. though not as pronouncedly as the tone There is more than the matter of the of the under parts. This contrast is not extent of the white feathering in the loral shown merely by a few examples but is and superciliary regions to distinguish observable in almost all of over one albiloris and superciliaris. In albiloris hundred and thirty skins, about evenly there is a marked seasonal difference in divided between the two areas. There the development of white in this region also is a certain average difference of size, (as elaborated by Van Rossem, 1931, Auk, although there is too much overlap to XLVIII, p. 35), but in superciliaris there permit this character to be used alone. is no such seasonal difference, although Twenty-one adult males from South there is a sexual difference of the same sort America have the tail 41-47 mm. in length in one of the conspecies of this form (cf. (av., 43.5); twenty-seven adult males discussion of P. pl. major). Albiloris is from Central America have the tail 36-43 somewhat larger than superciliaris. Ex- mm. (av., 40). Some South American cluding southwest-Mexican birds, of which examples have a certain amount of grayish more will be said below, twenty-three adult shading on the lateral under parts, but it males have the tail 42.1-48.2 mm. in is of a lighter tone than that of the Central length (av., 45.5). It is decidedly more American birds. In series, the distinctions purely whitish below, without the gray are pronounced. tones of superciliaris, or at least with much 1 Earlier papers in this series comprise American paler gray shading. The black areas at Museum Novitates, Nos. 500, 509, 523, 524, 538, 545, 558, 584, 646, 647, 668, 703, 728, 753, 756, 757, the bases of the outer rectrices are notice- 785, 819, 860, 861, 862, 889, 893, 894, 917, 930, 962, ably more extensive than in superciliaris 963, 994, 1042, 1043, 1044, 1045, 1066, 1095, 1108, 1109, 1126, 1127, 1159, and 1160. (most easily compared on the subexternal 2 AMERICAN MUSEUM NOVITATES [No. 1168 rectrix), and the pale outer border of the the largest of the series. The lengths of larger tertials is broader, more sharply tail in the adult males from different defined and often more purely white. The regions are as follows: W. Mexico (4 top of the head in the adult males is more skins), 48-51 mm. (av., 49.6); W. Honduras bluish and more highly glossy. (6), 44-48 (av., 46.1); C. Guatemala (5), These characters appear together in the 42-46 (av., 44.9); Nicaragua (5), 42-48 specimens examined, and I am unable to (av., 44.8); Costa Rica (7), 44-48.2 (av., find a single example that is not readily 45.9). assignable to one form or the other. Even Ridgway (1904, Bull. U. S. Nat. Mus., using only the broad, full superciliary of L, pt. 3, pp. 725, 726) restricted albiloris to superciliaris as a single criterion, there ap- Guatemala and Oaxaca, Mexico; bairdi, pears to be positive demarcation, and I to Nicaragua and Costa Rica. He found have seen no specimen of albiloris with male albiloris (all from Oaxaca) to have the more than a thin line above the orbit. tail 49-56 mm. (av., 52.6), but Van Ros- With this segregation of characters, I sem (loc. cit.) included all west-Mexican have specimens of both forms from Bebe- birds of this group in "bairdi" with an dero, Costa Rica, and the vicinities of average tail-length (males) of only 48 mm. Matagalpa and Chinandega, Nicaragua. Obviously there is much work still to be A series of albiloris from Progreso, Guate- done on the variations and distinctions in mala, is from the eastern side of the water- the albiloris group, and it appears probable shed, while the specimens of superciliaris that there is a large form in western Mexico from the same country are also from the still to be characterized and named. I do east, although nearer the coast and in a not see how it can be referred to "bairdi" less arid region than Progreso. Possibly or how a "bairdi," based on Nicaraguan the distributional segregation of the two birds, can be distinguished from albiloris. forms throughout their ranges is more a Returning to the consideration of P. matter of ecology than simple geography. plumbea bilineata, I must say that I have Nevertheless, the remainder of the two only one bird from Santa Marta to repre- series of these forms from northern Costa sent the north-Colombian population of Rica, Nicaragua, and Honduras are from the subspecies, in this case the population opposite sides of the cordillera. In south- to which the name bilineata adheres. This ern Costa Rica, superciliaris occupies both single bird agrees in coloration with the sides of the mountains. On the other west-Ecuadorian and west-Peruvian birds, hand, albiventris, which I believe is to be although in size it can be matched in both considered as a form of albiloris, is found these and Central American skins of in northern Yucatan and on Cozumel superciliaris. In view of the approxima- Island, on the Caribbean side of Mexico, tion in color, therefore, I follow earlier in arid regions. authors in assigning the west-Ecuadorian I am unable to follow Van Rossem's and west-Peruvian birds to bilineata in distinctions between albloris and "bairdi" spite of the hiatus in distribution that (1931, Auk, XLVIII, pp. 34-39). He occurs in western Colombia where the gives the length of tail in albiloris as 53 form has not been found to date. mm., in "bairdi" as 48, and restricts As to the matter of specific association albiloris to eastern Guatemala and south- with plumbea, the case appears to be fairly ern Mexico on the Atlantic side of the clear. A male of plumbiceps from Honda, mountains (but without giving Mexican Colombia, shows decided, white lores and localities for this latter form). The sup- superciliaries, and another male from the posed "bairdi," described by Ridgway same locality has a lesser amount of white from Nicaragua, is said to be confined to in the same area, at the tips of the feathers. the Pacific drainage. A female from the foot of Mt. Duida has a My only Mexican birds of this species, very slight suggestion of white above the except the type of albiventris, are from the lores, and one from the upper Orinoco has Pacific side of Oaxaca, and these birds are traces of white above the eyes. The 1942] STUDIES OF PERUVIAN BIRDS. XLII 3

amount of white on the outer three rec- that this close approximation of the sexes trices varies sufficiently to show perfect would be a sufficiently distinctive character transition between plumbiceps and bilineata to serve as a basis for specific isolation of in this respect. General measurements maior, but there is much evidence to show good intergradation. In distribu- weaken this idea. tion there is no conffict. In this series of bilineata, parvirostris, The conclusion to be reached from this anteocularis, plumbea, and plumbiceps there evidence is that bilineata and its accepted are occasional specimens dressed in typical conspecies must be added to the plumbea female plumage except for their noticeably group through demonstrable relationship darker heads. These birds were marked to plumbiceps. by their collectors as females, and although In Perd, the range of bilineata is re- these determinations have been questioned stricted to the Pacific coast and western in one or two cases by subsequent workers, slopes of the western Andes from Pacas- I believe that the birds in question actually mayo north to the Ecuadorian boundary, are females and not young males as sug- thence extending northward to Esmeraldas, gested. A similar bird is the type of Ecuador. Peruvian records are from Pacas- plumbiceps which was sexed by the col- mayo, Tembladera, Guadalupe, and Lechu- lector as a male but which is rather gal, as well as the localities mentioned in exactly matched by other skins sexed as the subjoined list of material examined. females. One female of anteocularis from Honda, Colombia, has the head with a Polioptila plumbea maior Hellmayr decidedly bluish sheen, and with the lores P(olioptila) nigricep8 maior HELLMAYR, 1900, and an abbreviated superciliary stripe Novit. Zool., VII; p. 538-Sueccha [= Succha], white. A female of plumbea from Faro, 3000 m., Perd ; d; Amer. Mus. Nat. Hist. Rio Jamunda, Brazil, is similar but has (Rothschild Collection). Polioptila bilineata andina HELLMAYR, 1903, much more prominent white about the Verh. k. k. zool.-bot. Ges. Wien, LIII, p. 224- eye though the top of the head is not quite Cajabamba, Perd; e f= 9 ]; Frankfort Mus. as dark as in the Honda skin of anteo- (Berlepsch Collection). cviaris. Likewise a female of parvirostris The collection of a good series of males from Puerto Indiana, Peru, has a lightly and females in the range once supposedly glossed cap and a suggestion of white above occupied conjointly by maior and andina the eye where there should be no white. has shown that these two forms are merely To prove conclusively that these birds opposite sexes of the same bird, which must are females and not males will require be known as maior. Dr. Hellmayr has particular studies in the field, but the examined the specimens and has reached evidence on which I base my supposition the same conclusions. is as follows. Several young males are at The discovery of this synonymy removes hand rather definitely ascertainable as one difficulty formerly in the way of such by reason of various fully black recognizing plumbea and bilineata as con- feathers which have appeared on the crown specific. It eliminates a supposed con- among the soft, drab remnants of the ffict in distribution, but it projects another juvenal plumage. There is no inter- problem into the field with regard to the mediate plumage of bluish gray or bluish- relationship of maior and plumbea. In tipped gray to be seen. The immature this case the difficulty lies entirely in the females have a similar juvenal plumage female plumage, for the males are not dis- with drab or dull gray pileum, and in the tinct enough from other members of the change to clearer gray cap there is no ap- plumbea group to cause much comment. parent intermediate dress. Various of the The females, however, are marked by a supposed dark-headed females show no steely black cap like that of the male but suggestion of immaturity, and in those relieved by a white stripe above the eye, which do show such suggestions, the thus rather closely resembling the males juvenal plumage may be clearly recognized of bilineata. At first sight it might seem as such. Evidently, therefore, this dark, 4 AMERICAN MUt SEUM NOVITA TES [No. 1168 but not coimpletely black, coloration of of the Amazon, but I believe that three the hea(l is not part of any immature specimens from near the mouth of the Rio plumage but belongs to the adult dress of Napo must be referred to it. Compared the female, with typical examples of which with examples of P. p. plumbea from vari- complete intermediacy is shown in the ous localities, the three birds at hand differ series at h-iand. by having the auriculars and lores more Thus, wlhile there is no complete inter- extensively black in the male, gray in the gradation (lemonstrable between the fe- female. There is no trace of white oni males of maior an(d the same sex of any the lores (as, indeed, is likewise true in other mnember of the plumbea group, there some examples of plumbea), and the is a strong tendency sporadically shown by auriculars have the white restricted to individlual variants. Consequently I place the lowermost border or the basal portion maior in the plumbea group where it oc- of the shafts on the lower border, although cupies an other-wise unoccupied space in there is a subocular space of white which the middkle of the Peruvian range of the reaches the lower margin of the orbit at species. P. p. maior is still so little the middle. Also, plumbea has the white knowin that lpossible variations in its border of the tertials and innermost plumage on the outskirts of its range have secondar-ies averaging broader than is yet to be dliscoveredl. Although the species shown by the three parvirostris, but some pirobably is restricted to the Tropical Zone examples are not distinguishable in this in Peru', this subspecies almost forms an respect. exception since it ascen(ls to an elevation Dr. Hellmayr writes me that the dif- of 9000 feet according to the (lata ascribed ference I have noted in the extent of black to the type. Otlher examples are from on the auriculars is shown by the type of lower elevations, dlown to 1500 feet parvirostris from Chamicuros and also by (Sauces, Rio Chamaya), which is definitely three males, one each from Moyobamba, tropical though in the upper level of the Tarapoto, and "Juanfu6," Huallaga [= Tropical Zone. The elevation of 9000 feet Juanjuui]. This character, therefore, ap- is sufficient to carry maior over the Por- pears to be reliable as a subspecific cri- culla Pass near Huancabamba to the west- terion for separating parvirostris an(l ern slopes of the western Andes, but all plumbea. specimens from the Pacific side are refer- The female, apparently undescribed to able to bilineata. (late, is like the male in almost all respects In a sense, therefore, maior is both a except that the top and sides of the head geographical and zonal representative of are gray, of a shade distinctly darker than the group, since neither bilineata nor the back. One of the two females at hand parvirostris ireaches such high elevations. is as liglht gray on the back as the male, This and the comparatively efficient but the other, from Puerto Indiana, is isolation are responsible, no doubt, for much darker, with the top of the head much of the unusual degree of differentia- even more obscure. Differences of about tion now observable in the form uinder the same degree are shown between in- (liscussion. dividuals of either sex in the series of Records of maior are froin Guajango, plumbea. Callacate, and Rio Chota, as well as How far to the eastward parvirostris ex- localities from wlhich material has been ex- tends I am unable to say. The specimeil amine(l according to the subjoined list. at hand from Anayacu represents the most eastern locality known to me. On the Polioptila plumbea parvirostris Sharpe other hand, plumbea apparently extends Polioptila parvirostris SHARPE, 1885, Cat. westward along the Amazon only to the Birds Brit. Mus., X, pp. 441, 448-Chamieuros, Jamunda on the north bank and a short R. Amazon; &; British Mus. distance west of the Tapajoz on the south I have seen no topotypical material, an(d bank. Actual junction of the ranges of this form has not been recorded from north plumbea and parvirostris is evidently non- 1942] STUDIES OF PERUVIAN BIRDS. XLII 5 existent, and a wide extent of territory on Villa Bella Imperatriz, 6 e, 1 "9 ," 3 9; Rio Tapajoz, Urucuritiba, 1 9; both banks of the Amazon is left without Rio Xing6, Porto de Moz, 1 d; any known member of this group. Rio Tocantins, Baigo, 2 e; Records of parvirostris, which is con- Mocajuba, 1 9. fined to Perd so far as known, are from P. p. innotata.- BRAZIL: Tarapoto, Chamicuros, Moyobamba, Juan- Rio Surumu, Frechal, 1 a, 3 9; juf, and Sarayacu, and the localities where Rio Branco, Caracarahy, 1 a. the specimens listed below were collected. BRITISH GUIANA: I believe that birds from southern Annai, 2 a, 2 9; Rupuruni River, 1 9. Venezuela belong with innotata rather than VENEZUELA: with plumbiceps. A series of nearly forty San Fernando de Atabapo, 1 a; skins from the Orinoco and Mt. Duida Caicara, 1 9; shows a distinct average of less heavy El Llagual, 1 9; Maripa, 4 cd; shading on the chest and sides than the Maipures, 1 d; more northern examples, agreeing with Agua Salada de Ciudad Bolivar, 1 a; specimens from British Guiana and north- Ciudad Bolivar, 5 a, 5 9; ern Brazil. There is some of Altagracia, 3 i, 2 9; overlapping Lalaja, 1 c, 1 9; characters whichever way the separation is Ayacucho, 2 e, 1 9; made, but a more satisfactory arrange- Esmeraldas, 1 a, 3 9. ment, from both taxonomic and distribu- P. p. ptumbiceps.- tional is the method VENEZUELA: aspects, proposed. (no other locality), 2 e(inlc. type); The type of plumbiceps Lawrence bears Tucacas, Estado Falcon, 1 d; no indicated locality other than "Vene- La Vela de Coro, 1 9; zuela." It was collected by S. C. Nash El Cuji, Estado Lara, 1 ci; who also collected the type of Myiarchus Cardpano, 1 cd; San Felix, Cuman&, 6 a, 2 9; ferox venezuelensis, described by Lawrence Los Dos Rios, 1 d; in the same paper with the Polioptita. La Guaira, 1 d; This Myiarchus is confined to northwestern Puerto Cabello, 1 (?). Venezuela in the neighborhood of the P. p. anteoeularis.- COLOMBIA: State of Carabobo where Nash's other "BogotM," 5 [ci]; specimens are likely also to have been Honda, 3 a, 1 9; obtained. Since the form in question is Chicoral, 3 e,1 9. known to occur at Tucacas, FalIc6n, not P. p. daguae. COLOMBIA: far from Carabobo, it is appropriate to Los Cisneros, 1 e (type); select Tucacas as type locality for Poliop- Rio Dagua, 1 9; tila plumbiceps Lawrence, 1865. Primavera, 1 d; "Yuntas" [= Juntas de Taman6], 1 9. P. p. bilinaa.- SPECIMENS EXAMINED PER16: P. p. ptumbea. Palambla, 1 d; DUTCH GUIANA: Paletillas, 2 cI, 2 9; Paramaribo, 5 ", 1 9; Pilares, 1 9; Kwata, 2 9. Samate, 2 e; FRENCH GUIANA: Chilaco, 1 e, 1 (?); Cayenne, 8 e, 3 9, 1 (?); Sullana, 5 c, 2 9; Approuague, 1 9; Lamor, 4 cI, 2 (?); Roche Marie, 1 9. Tumbes, 1 a; "BRITIsH GUIANA:" (errore ?)1, 1 9. Tembladera, 2 9. BRAZIL: ECUADOR: Faro, 15 e, 8 9; Duran, 4 a; Park, Flor de Prado, 1 9; Daule, 1 d; Maraj6, 1 d; Esmeraldas, 4 a, 4 9; 1 1 1 Locality probably erroneous. The bird was col- Manta, e, 9; lected by Alexander and, like various other of his Bahia (Manavl), 1 a", 1 9; skins, is the only evidence for including the form to Isla La Plata, 2 e, 2 9; which it belongs in the British Guianan avifauna. All other records of the present species from this Guayaquil, 2 a, 3 9; country belong to P. p. innotata. Nanegal, 1 a"; 6 AMERICAN MUSEUM NOVITATES [No. 1168

Bahia de Caraques, 1 9; COSTA RICA: Santa Elena, 1 d; (Bebedero, Las Cafias, and Bagaces), 7 c, Chongocito, 2 ci, 1 9; 3 9. Isla de Puna, 2 e, 1 9; NICARAGUA: Rio Pullango, 1 "; (Matagalpa, Leon, Calabasas, VolcAn de Casanga, 1 "; Chinandega, 4 miles north of Chinandega, Rio Jubones, 1 d; Matagalpa, San Rafael del Norte, Chongon Hills, 1 9; Corinto, Savana Grande near Managua, Portovelo, 4 ", 1 9; Calabasas, south of Metapa), 6 c, 6 9. Bucay, 1 6. GUATEMALA: COLOMBIA: Progreso, 7 c, 1 9, 3 (?).

Bonda, Santa Marta, 1 9. MEXICO: I P. p. superciiari&.- Tehuantepec, Tapana, 1 9; PANAMA: Santa Efigenia, 2 "; (PanamA R. R.), 1 , 1 9 (cotypes) Oaxaca, 2 d. Almirante, 4 , 2 9, 1 "9"; P. a. albivenris.- La Marea, 1 "; MEXICO: El Virlano, 1 d; Yucatan, Temax, 1 e (type). Cituro, 1 9; Chiriqui, 2 e, 1 9; La Chorrera, 1 d; THE POLIOPTILA GUIANENSIS GROUP Coiba Is., 1 d; Santa Fe, Veraguas, 1 "; During the study of Polioptila plumbea, Cerro Largo, 1 d; the congeneric species, guianen8is, was "PanamA," 2 e, 1 9. examined for comparison. It was found COSTA RICA: at a Miravalles, 1 d; that the material hand included very Carrillo, 2 e; distinct new subspecies very nearly linking Bebedero, 1 e, 1 9; the allied schistaceigula with it. Orotina, 1 c; In spite of the purely extralimital char- Boruca, 2 e, 1 9; acter of this group, I include the discussion El Zapotal, 5 e, 2 9; Aquinares, 1 d; of it here where it may not be out of place Lim6n, 1 d; beside the treatment of other members of Volcan de Oso, 1 c; the genus. The description of the new El Pozo, Rio T6rraba, 1 c, 1 9. NICARAGUA: form is as follows. Tuma, Matagalpa, 1 cd; Rio Grande, 1 d; Polioptila guianensis facilis, Los Sabalos, 2 9; Chinandega, 1 9; new subspecies Chontales, 1 9. TYPE from Solano, Rio Cassiquiare, Venezuela. GUATEMALA: No. 433,542 bis, American Museum of Natural Chimoxan, 2 c, 1 9; History. Adult male collected May 5, 1929, by Vera Paz, 1 c; the Olalla brothers. Finca Chama, 1 e,2 9; DIAGNoSIs.-Similar to P. g. guianensi of the Secanquim, 1 c, 1 9; Guianas, but eyelids without pronounced white; "Guatemala," 2 c, 1 (?). tail with considerably less white; all the rec- HONDURAS: trices with black bases reaching nearly to or El Boqueron, 1 d. beyond the tips of the under tail-coverts; P. p. maior.- penultimate pair with about the basal half PERU: black; third pair with only a relatively narrow Succha, 1 6 (type); white tip; breast somewhat paler gray. Vifia, 1 d; RANGE.-RiO Cassiquiare, Venezuela, and Cajabamba, 3 c, 2 9; upper Rio Negro, Brazil. Araqueda, 1 (?); DESCRIPTION OF TYPE.-Upper parts Neutral Pucara, 1 d; Gray X Deep Neutral Gray; upper tail-coverts Huancabamba, 1 e, 5 9; darker with more blackish centers. Lores and Perico, 2 , 2 9, 1 "9"; malar region Light Neutral Gray, auriculars a Sondorillo, 1 9; little darker; lower eyelid only faintly whitish, Sauces, 1 9. not much paler than the lores; chin and throat P. p. parviro8tris.- whitish, breast and sides Pale Neutral Gray; PER-6: belly, flanks and under tail-coverts white; Puerto Indiana, 1 c, 1 9; thighs gray; wings blackish; primaries, except Anayacu, 1 9. for two outermost, externally narrowly mar- P. a. albiloris. gined with light gray; secondaries with margins 19421 STUDIES OF PERUVIAN BIRDS. XLII 7 broader and a little paler; tertials with exposed in this respect facilis is nearer guianensis, surface largely gray; upper wing-coverts with exposed edges Deep Gull Gray; under wing- although a little lighter in tone. coverts whitish; inner margins of remiges white; P. schistaceigula of northwestern Ecuador alula black, with narrow pale gray outer mar- and western Colombia certainly is closely gins. Tail with outer rectrices white except for related to guianensis, if it is not conspecific. narrow blackish bases not quite hidden by the under tail-coverts; second pair with broader As found in Colombia, it presents a further black bases, the line of demarcation from the extension of the white facial markings white terminal area starting from a point 30 mm. exhibited by the British Guianan skin of from the tip of the outer web, crossing the shaft guianensis and has an even greater reduc- 22 mm. from the tip, and reaching the inner than the margin about 12 mm. from the tip; third pair tion of white on the rectrices with an irregularly shaped white tip 4 mm. long specimens of g. facilis. The type (a male) (at shaft) on outer web, 71/2 mm. on inner web; is very sooty on the breast and has most remainder of tail black with only very faintly of the sides of the face gray, with very suggested grayish tips at the extreme apex of upper and the fourth pair. Bill with maxilla dull blackish narrow white borders on the (in dried skin), mandible a little paler; feet lower eyelids and a short white line over blackish. Wing, 461/2 mm.; tail, 441/2; ex- the anterior auriculars. The two females posed culmen, 9; culmen from base, 121/2; at hand from Colombia (Rfo Dagua and tarsus, 16. Rfo Cauca) have the whole side of the REIMARKS.-Female like the male but head white, including a relatively broad paler on the breast and sides. The exact superciliary stripe, except for a restricted extent of black at the bases and of white dusky anteocular spot and a dark postocu- at the tips of the rectrices varies in the two lar stripe. It is possible, therefore, that males, but the arrangement is of the same the Colombian examples are not true general style, with most of the outer schistaceigula, but it is also possible that feathers, a little more than the terminal their differences from the Ecuadorian bird half of the next pair, and a small terminal are sexual or individual. portion of the next white, with a faint sug- Future material from various localities gestion of white or pale gray on the next, may make it desirable to place guianensis and with the two middle pairs entirely and facilis with schistaceigula, but at pres- blackish. This is in decided contrast to ent it will be sufficient to note that they the tail of guianensis which has the three are more closely related to each other than outer rectrices almost entirely white. to any of their congeners. Much more A pair of birds from Caxiricatuba, Rio material will be necessary to determine the Tapajoz, Brazil, appear to belong to P. g. limits of individual and racial variations. paraensis (Todd, 1937, Ann. Carnegie Mus., XXV, p. 255-Benevides, Para; SPECIMENS EXAMINED, c1; Carnegie Mus.). They are even P. g. guianens8.- paler than facilis in comparable plumage, FRENCH GUIANA: Pied Saut, 1 S. although the male of paramnsis is rather BRITSH GUIANA: similar below to the female of facilis. The Potaro Landing, 1 9. female is noticeably paler both above and P. g. faclis.- below. Both birds have a fine, white line VENEZUELA: Solano, Rio Cassiquiare, 1 e (type); occupying the eyelids, quite inconspicuous Rlo Pescada, Mt. Duida, 1 e. in comparison with the facial markings of BRAZIL: guianernsis but noticeable in comparison Mt. Curycuryari (foot), Rio Negro, 1 9. with facilis. The tail apparently is longer P. g. para5nsis.- BRAZIL: than in facilis or guianensis, being 48 mm. Caxiricatuba, Rio Tapajoz, 1 i, 1 9. in the type, according to Todd, and the P. schi8taceigula.- same in the Caxiricatuba male; 46 in the ECUADOR: Caxiricatuba female. The upper parts Cachabi, 1 e (type). COLOMBIA: have a slight mouse-gray tinge in compari- Puerto Valdivia, Rio Cauca, 1 9; son with the bluer gray of guianensis, and San Jose, Rio Dagua, 1 9.