The Condor96692-702 0 The Cooper Omitholo&al Society 1994

SPECIES STATUS OF THE WHITE-FRONTED , SERENA (PIPRIDAE), WITH COMMENTS ON CONSERVATION BIOLOGY ’

&CHARD 0. PRUM Museum of Natural History and Department of Systematicsand Ecology, Universityof Kansas, Lawrence,KS 66045 USA

Abstract. The White-fronted Manakin, Lepidothrix serena(formerly in the genusPipra), is currently recognizedas a polytypic speciesthat includesnominate serena,from the eastern Guianan region, and suavissima,from southern and eastern Venezuela and Guyana. Nom- inate serena and suavissimaare significantly different in , syringeal morphology, and vocalizations. The distribution of the two forms has not been completely documented, but no contact between the populations is known. Derived morphological and behavioral novelties indicate that the two forms are sister taxa, and that they constitute distinct phy- logeneticand biological .The recommended common name of the newly recognized Lepidothrix suavissimais the Tepui Manakin. Maintenance of the current biological species may underestimate the diversity of speciesin the Neotropics and hamper the documentation of diversity in the Neotropics that is important to conservation biology. Key words: Pipridae;J_epidothrix serena; Lepidothrix suavissima;species limits; syringeal morphology;vocalization; Neotropics: conservation biology.

INTRODUCTION tiation of Neotropical demon- The current specieslimits of many Neotropical strate that many isolated populations that are birds are based on taxonomic decisions made by recognized as subspecies,or not recognized as ornithologists in the early and mid-twentieth separate taxa at all, are much more genetically century. Many polytypic speciesof Neotropical differentiated than many biological species of birds were established on the basis of plumage Nearctic (Capparella 1988, 199 1; charactersalone in complete absence of natural Hackett and Rosenberg 1990; Hackett 1993). history information (e.g., Ridgway 1907, Hell- These studiesstrongly suggestthat there are many mayr 1929, Meyer de Schauensee 1966). Most more speciesof Neotropical birds than currently of thesetaxa are currently considered to be valid recognized. This underestimate of Neotropical biological species,even though the specieslimits diversity could lead to inaccuracies in analyses among the differentiated forms within them may of regional endemism and the reconstruction of never have received detailed consideration. Re- the biogeographic history of these regions that evaluation of polytypic Neotropical taxa has re- are critical to conservation biology. Reanalysis vealed many highly distinct forms that are es- of variation in plumage, morphometrics, other sentially hidden within wide-ranging polytypic aspectsof morphology, molecular characters,and species.For example, Robbins and Ridgely (1992) behavior of many Neotropical birds is required have shown that the isolated Choco endemic to re-evaluate the status of these wide-ranging, Nyctiphrynus rosenbergiis highly differentiated polytypic biological species. in song,plumage, and mass from the Amazonian Here, I address the specieslimits of two dif- N. ocellatus, with which it was formerly com- ferentiated basal taxa of (Pipridae). bined in a “biological” species. Indeed, rosen- Lepidothrix serena is currently recognized as a bergi is probably not even as closely related to polytypic speciescomposed of two allopatrically ocellatus as are the distinct Central American or parapatrically isolated populations in north- speciesyucatanicus and mcleodii (Robbins and ern South America. The nominate form of serena Ridgely 1992). Analyses of molecular differen- was originally describedby Linnaeus (1766) from Cayenne material, and placed in the genusPipra. A second taxon was subsequently described as 1Received 14 October 1993. Accepted 24 February suavissimaby Salvin and Godman (1882) 1994. based on specimens from Roraima and the Me-

[6921 SPECIESSTATUS OF THE WHITE-FRONTED MANAIUN 693 rume Mountains in then British Guiana. The two iodine-stained syringeal specimens of both spe- forms were treated as separate speciesby Hell- cies were observed and illustrated using a Wild mayr (1906) and Chubb (1921), but they were M5 stereo dissection microscope (serena: ROM combined as subspeciesof Pipra serena without 127643, USNM 515106, 515108, 515109, comment by Hellmayr (19 10, 1929). No com- 515113,515114;suavissima:AMNH9366,9368, ments in the literature were made on the system- 10376,10377,816768,BM-A-1968-46-19).Tape atics of serenauntil Haffer (1970,1974) and Snow recordings of vocalizations were collected from (1975, 1979) proposed that the polytypic serena my own field recordings,recordings by Tom Da- be included in a species-group with coronata, vis, and from recordingsby Paul Schwartz,Theo- nattereri, iris, vilasboasi,coeruleocapilla, and is- dore A. Parker III, and Mario Cohn-Haft de- idorei. Prum (1988) proposeda phylogeny of the posited at the Cornell Laboratory of Natural monophyletic serena species group based on Sounds (LNS). Analyses of vocalizations were plumage traits in which nominate serena and done using Canary 1.1 computer software pro- suavissimawere hypothesized to be sister taxa. duced by the Cornell Laboratory of Natural In a phylogenetic analysis of the entire Pipri- Sounds. Sonogram figures were prepared using dae, I uncovered no evidence of relationship be- Kay Elemetrics Vibralyzer 7030-A. tween the serena species group and the rest of genusPipra. Basedon syringeal morphology, the RESULTS serena speciesgroup is more closely related to a group including , , and An- DISTRIBUTION tilophia. Given evidence of the polyphyly of Pi- The nominate form of serenais distributed from pra (sensu Hellmayr 1929, Snow 1979), I re- the Acary Mountains (in disputed territory along moved the serena speciesgroup from Pipra and the southern border between Guyana and Suri- placed it in its own genus, for which the name name), to the interior of Suriname and French Lepidothrix Bonaparte was available (Prum Guiana, south to near Manaus, and south- 1992). em Amapl near the mouth of the Amazon River Within the polytypic speciesLepidothrix sere- (Fig. 1). Snow (1979) overlooked the Acary na, I discovered a tremendous difference in sy- Mountain record (Blake 1950) and several Bra- ringeal structure between the two subspeciesthat zilian localities (Willis 1977, Novaes 1978) in is greater than that found between many families his description of the range of nominate serena. of oscine birds (Prum 1990a, 1992). As a result, Several other localities have been published sub- I have analyzed the variation in the distribution, sequently that further established the range of plumage, size, syringeal morphology, and song serenain northern Amazonian Brazil (Willis and of nominate serenaand suavissimato re-evaluate Oniki 1988, Stotz and Bierregaard 1989, Bier- the status of the two taxa. regaard 1990). Large portions of the presumed range of serena have not yet been documented. Based on current information, serena likely oc- METHODS curs in portions of western and southern Suri- Study skins and spirit specimenswere observed name, and northern Para, Brazil between Ma- from the collections of the American Museum naus and Amapd. In Suriname, serena occurs in of Natural History (AMNH), the British Muse- primary forest above 200 m in altitude, and is um (Natural History) (BM), the University of not found in coastal forest (Haverschmidt 1968, Kansas Museum of Natural History (KU), the Prum 1985). Royal Ontario Museum of Natural History Populations of suavissimahave been recorded (ROM), and the United StatesNational Museum throughout the tepui highlands of southern Ven- of Natural History (USNM). Distributions of se- ezuela (just reaching into Brazil at the southern rena and suavissimawere plotted from published tip of Venezuela), in northeastern Venezuela, and records and from localities on museum speci- in western, central, and eastern Guyana (Fig. 1). mens. Plumage was described from museum There are no records of suavissima from the study skins and compared to a standard color southern third of Guyana or the entire Rio Bran- reference (Smithe 1975). Tarsus, flat wing, and co drainage, Brazil. This lack of records may tail were measured on 60 specimens of serena reflect the actual distribution. Moskovits et al. and suavissima.Twelve cleared-and-stained,and (1985) did record suavissima at the Ecological 694 RICHARD 0. PRUM

65” 60” 55” 50” FIGURE 1. Map of the distributionof nominateserena and suavissima.Circles are serenalocalities. Squares are suavissimalocalities.

Station of Maraca on the Uraricoera River, Ro- The closest localities in the known distribu- raima, Brazil (03”25’N, 6 l”4O’W). However, Stotz tions of the two taxa come along the lower Cou- (1993) recordeda hybrid between suavissimaand rantyne River, which forms a portion of the bor- Lepidothrix coronata in northern Roraima, Bra- der between Guyana and Suriname. In eastern zil (04”29’N, 61”09’W) about 50 km from the Guyana, suavissimais known from Tiger Creek, closestknown locality at Pauri-tepui in southern a small west or left bank tributary of the Couran- Venezuela. The distribution of suavissimamay tyne (Chubb 192 1, Stephensand Traylor 1985). extend a little further into Roraima, Brazil than Nominate serenais known from severallocalities is currently known, but it is probably not gen- in central and eastern Suriname, but no records erally distributed south of current localities in are available from western Suriname along the the Rio Branco drainage or to the Rio Negro. In lower Courantyne. However, serena is known Venezuela, suavissima has been recorded be- from the Acary Mountains which are between tween 500-1,800 m (Meyer de Schauenseeand the left bank of the upper Courantyne, and the Phelps 1978) and would appear to be restricted east or right bank of the New River. The known to tepuis, but old collections in central Guyana distributions of serena and suavissimaare allo- indicate that suavissimamay range down to near panic. It is possible that serena and suavissima sea level. come into parapatric contact somewhere on the SPECIES STATUS OF THE WHITE-FRONTED MANAIUN 695

TABLE 1. Sexualdimorphism in tarsus,flat wing, and tail in nominateserena and suavissima.See text for details.

serena Male 12 16.2 0.47 13 52.2 0.83 12 26.9 2.18 Female 12 15.5 0.72 11 54.2 1.74 11 27.6 1.38 Significance(t-test) P = 0.015 P = 0.0055 P = 0.34 suavissima Male 22 15.2 0.48 21 57.9 1.07 22 28.3 1.07 Female 13 15.1 0.41 13 58.6 1.04 13 29.8 1.01 Significance(t-test) P = 0.41 P = 0.048 P = 0.0002

lower Courantyne or New Rivers. However, repeated in an illustration of suavissima by G. nominate serenaand suavissimaboth prefer hu- Tudor in Meyer de Schauenseeand Phelps(1978), mid forests above 200 m in altitude, and their and by Prum (1988) in a phylogenetic analysis rangesmay be separatedfrom one another along of the serena species group. Data from recent the lower Courantyne by inappropriate white specimens with soft part colors described sand Savannah-woodland or humid forests be- (USNM), and photographs from Suriname (R. low 200 m. 0. Prum, VIREO) and near Manaus (VIREO) document that the iris color of all populations is PLUMAGE brown. The plumage of serena and suavissimaare sim- ilar, but both sexesof each form have distinctive, SIZE fixed differences in plumage coloration. (Capi- Both serena and suavissima exhibit significant talized color names refer to Smithe [ 19751).Males in size (Table 1). In serena, of serena and suavissima are velvety black on males were significantly smaller than females in the head, back, wings, tail, and upper breast, and wing length (t-test, P 5 O.OOSS),but significantly have a white forecrown, a blue rump (Venetian larger in tarsus length (t-test, P 5 0.0 15). Males Blue), and a yellow belly. In male serena, the and females were not different in mean tail length belly is Spectrum Yellow, and a small patch on (t-test, P I 0.34). In suavissima,males were sig- the upper breast is Orange Yellow; the forecrown nificantly smaller than females in wing (t-test, P is white with a slight Sky Blue tinge in very few I 0.048) and tail (t-test, P 5 0.0002) lengths, specimens(illustrated in Haverschmidt 1968). In but were similar in tarsus length. male suavissima,the belly is Orange Yellow, the In general, serena is slightly smaller than sua- upper breast is entirely black, and the trailing vissima,and males are smaller than females. The edge of the forecrown is Sky Blue (illustrated in exception is in tarsus length in serena which is Meyer de Schauenseeand Phelps 1978). No in- bigger in males. Although many of these differ- termediates in male plumage were observed, and encesare statisticallysignificant, they are not great none has been described. enough to make individuals of each taxon diag- Female serenaand suavissimaare both largely nosably different in size. grassgreen above. Female serenaare light sulfur yellow below. Female suavissima are brighter SYRINGEAL MORPHOLOGY yellow below, and have a distinct blue tinge on Prum (1992) describedthe syringeal morphology the forecrown. of serena and suavissima in detail; however, a Both sexesof both forms have black legs,black larger sample including five more syringeal spec- bills, and brown irises. The iris of serena was imens of serena was examined for this analysis. erroneously illustrated as white by P. Barruel in Syringeal terminology and variation in the sy- Haverschmidt (1968); the artist apparently ringeal morphology of manakins is described in matched the iris color to the white of Pipra au- Prum (1992). The syrinx of suavissimais similar reola and Pipra eythrocephala. This error was to other speciesof Lepidothrix except that it is 696 RICHARD 0. PRUM slightly larger in diameter, has an extended series territorial males and are occasionally inter- of dorsally cartilaginous tracheal A elements, and spersedwith low, throaty, whistled “boop” notes a few double, medially cartilaginous, complete (Fig. 3B). During display, males also give a soft, A elements (Figs. 2A, B). The syrinx of serena descending “puurr” note (Fig. 3C). The whree has all of the features of suavissimawith an ad- and boop calls are essentially the same on tape ditional suite of derived features that are unique recordings from near Manaus. among all manakins (Figs. 2C, D). All of the The vocalizations of suavissima have been unique featuresof the syrinx of serenaare related previously described as a “squeak” by Meyer de to its increase in size. The diameter of the syrinx Schauenseeand Phelps (1978). Based on record- of serena is 3.8 mm at Al (n = 7; SD = 0.1 l), ings, the main vocalization given by territorial and 2.8 mm at A10 (n = 7; SD = 0.13). The males is a sharp, nasal, slightly rising “aank” diameter of the syrinx of suavissimais 2.6 mm (Fig. 3D). Males also give a rapid, emphatic, pip- at Al (n = 6; SD = 0.13), and 1.6 at A10 (n = ing series of 7 or 8 notes which rise and fall in 6; SD = 0.05). The mean diameters of Al and pitch and emphasis:“ whee-pee-pee-.. .-pee” (Fig. A6 of nominate serena and suavissimadiffer by 3E). This piping call was recordedfollowing play- more than nine standard deviations, and are sig- back of the aank call and during counter-singing nificantly different at the P < 0.01 level. Asso- among males. ciated with the widening of the syrinx, the tra- The territorial calls of serena and suavissima chea in serena is twisted and distorted in shape are similar in note structure and behavioral con- into an oval, in cross section. The bronchi are text. These calls are apparently homologs that also expanded in size, and the B elements are have diverged since common ancestry between further distorted in shape.Further, the slight me- the two taxa. Sonograms of these vocalizations dial, cartilaginous connections among the bron- reveal that the whree calls are almost twice as chial A elements in suavissimaare expanded in long as aank calls (Figs. 3A, D; whree: n = 13, serenainto a large cartilaginous plate that is con- X = 253 msec, SD = 27; sank: n = 12, X = 136 nected to the dorsal and ventral ends of four or msec, SD = 22), and that this difference is sta- five bronchial A elements, and forms a cartilag- tistically significant (t-test, P -c 0.01). The two inous plate that is the medial surfaceof the bron- vocalizations also differ in frequency and fre- chi (Fig. 2C). Because of the distortion of the quency modulation. The whreecall of serenahas bronchial diameter, this cartilaginous plate is fre- a primary frequency band between 1.5 and 2.2 quently buckled or depressedinto the bronchial kHz. Recordingsof the call sometimes show oth- lumen. In serena, the series of dorsally cartilag- er weaker harmonic bands near 4 and 6 kHz inous tracheal A elements extends at least to A 15 (Prum 1985), but under other conditions no but can continue to A19. All species of Lepi- identifiable harmonics are recorded. The whree dothrix lack intrinsic syringeal muscles, and the call either rises slightly in pitch, from 1.5-l .8 M. tracheolateralis of serena inserts on A 1 as in kHz to 2.0-2.2 kHz, or it remains centered on a other speciesof the genus. single band of frequencies. All whree calls show distinctive, rapid oscillations between the max- SONG imum and minimum frequenciesover very short Recordings of serenacome from the Brownsberg time intervals. These frequency modulations give Nature Park, Suriname (04”55’N, 55”12’W) [tape the call its throaty, rolling quality. recordings by R. 0. Pi-urn, and T. Davis], and In contrast, the aank call of suavissimais com- near Manaus, Brazil (02”4O’S, 6O”OO’W) [LNS posed of a main frequency band that begins near 487431. The available recordings of suavissima 1.5-1.6 kHz and rises rapidly over a few msec were made in Rio Grande, eastern Venezuela to form a wide band between 2.2 and 2.4 kHz (07”58’N, 61”53’W) [P. Schwartz recordings, for the last two thirds of the note (Fig. 3D). Many housed at LNS], and on La Escalera, in south- recordings show weaker harmonic bands at 1.2, eastern Venezuela (06”00’N, 61”lO’W) [LNS 3.6,4.8, and 6.0 kHz. The aank call showsnone 304461. of the rapid frequency modulations found in the I described the song of serena in Suriname as whree call. a soft, throaty, rolling “whree” (Fig. 3A), with The other vocalizations of serena and suavis- the quality of a toy police-whistle (Prum 1985). sima are apparently unique to each taxon, and The whree notes are given in a long series by not homologous to any vocalizations in the oth- SPECIES STATUS OF THE WHITE-FRONTED MANAKIN 697

FIGURE 2. Syringealmorphology of nominate serenaand suavissima.(A) Dorsal and (B) ventral views of the syrinx of suavissimu(AMNH 816768, 9366 respectively); (C) Dorsal and (D) ventral views of the syrinx of nominate serenu(ROM 127643). See text for descriptions. Al indicates the Al syringealsupporting element. 698 RICHARD 0. PRUM

6- A

4- kHz -

I 600

6- D

I I , 200 400 600

FIGURE 3. Sonogramsof vocalizations of nominate serena: (A) two examples of the whree call, (IS) the hoop call, and (C) the purr call. Sonowams of vocalizations of suavissima: (D) two examples of the aank call, and (E) the piping cab. See text for dkiptions. er’s repertoire. The soft boop call of serena is a DISCUSSION short note (- 100 msec) that rises slightly from 0.8 to 1.0 kHz (Fig. 3B). The soft, descending Nominate serena and suavissima differ signifi- puurr that is given during display appears as a cantly in many aspects of plumage, syringeal seriesof 10 to 12 short notes that begin near 1.5 morphology, and song. The plumage differences kHz and descend to near 1 kHz over 250-300 between the two forms are found in both males msec (Fig. 3C). and females, and are fixed among populations. The agonistic, piping trill given by suavissima The syringeal morphology of the two taxa is tre- is a seriesof seven or eight notes (Fig. 3E). Each mendously different. The syrinx of nominate se- note rises rapidly in frequency over 50-55 msec rena has a suite of unique, derived syringeal fea- and then drops back to the initial frequency in tures which constitutea major changein syringeal - 15 msec. The first and last notes in the series organization. This degree of syringeal differen- have initial and maximum frequency ranges of tiation is unknown previously among any avian 2.3-3.5 kHz. Each note in the seriesincreases in congeners,and certainly not among conspecifics frequency range until the middle notes, which (Prum 1992). Consequently, the two taxa differ vary from 2.64.3 kHz, and then tapers off to significantly in the acoustic structure of their vo- resemble the values of the initial notes. In some calizations. Despite striking differencesin syrin- recordings, these calls show related harmonic geal morphology, the main territorial calls of the bands at 6-8 kHz and above 10 kHz. Interest- two taxa are apparently homologous, and differ ingly, in all three available recordings, male sua- in length and frequency and frequency modu- vissimaalternated strictly between sevenand eight lation. Each taxon has additional vocalizations note trills. that have no apparent homolog in the other’s SPECIES STATUS OF THE WHITE-FRONTED MANAKIN 699

vocal repertoire. Since most vocalizations of sub- ties of manakins (R. 0. Prum, pers. observ.). oscine passerinesare considered to be innate and Vocal differentiation has been used as evidence not learned (Kroodsma 1984) the vocal differ- of speciesdifferences in tyrant flycatchers (Ty- ences between the two forms also constitute ev- rannidae) in which songsare innate (e.g., Lanyon idence of genetic differentiation rather than 1978, Johnston 1980). Since territorial vocali- merely of cultural divergence. The two taxa also zations and male plumage are used in mate at- differ slightly in size. traction displays in manakins, it is likely that These data support the conclusion that serena differentiation in these traits is directly related and suavissima are distinct, basal, diagnosable to mate preferencesthat would be maintained in lineages (Cracraft 1983, McKitrick and Zink secondarycontact. In conclusion, differentiation 1988) with independent evolutionary histories in plumage, size, syringeal morphology, and vo- and fates (Wiley 1978, Frost and Hillis 1990). calizations between serena and suavissima are On this basis, I would recommend that the two significant, comparable or in excessof the level taxa be recognized as separate phylogenetic, or found in other recognized biological species of evolutionary species(Wiley 1978, Cracraft 1983, manakins and tyrant flycatchers, and likely to McKitrick and Zink 1988, Frost and Hillis 1990). result in reproductive isolation if these taxa were These data also support the conclusion that in secondary contact. For these reasons, I rec- these taxa are distinct biological species.In ap- ommend that serena and suavissima be recog- plying the biological speciesconcept to these al- nized as separatebiological species. lopatric or parapatric taxa, the degree of differ- An appropriate common name for the newly entiation between serena and suavissima must recognized speciesLepidothrix suavissimais the be gauged as an indication of the likelihood of Tepui Manakin, based on its distribution in this extensive interbreeding were they to come in distinctive Neotropical land formation. The contact with one another. One way to address monotypic Lepidothrix serena should continue this issue is to compare the degree of differen- to be called the White-fronted Manakin. tiation to other currently recognized biological Further field research in western Suriname, speciesof manakins and other suboscines.The eastern Guyana, and the Rio Branco and Rio degreeof plumage differentiation between serena Jauaperi drainages, Brazil, could further eluci- and suavissimais similar to speciesof the Pipra date the limits of the distributions of Lepidothrix eythrocephala clade, the Pipra aureola clade, and serena and L. suavissima. At present the lower Chiroxiphia (Pi-urn 1992). Unfortunately, vari- Courantyne River and the New River appear to ations in the application of the biological species delineate the northern border of the distribution concept among authors make other comparisons of the two forms. The limits of their distribution unclear. Although the highly differentiated spe- in Brazil are even lesswell known. It is probable, cies of Manacus hybridize along certain limited however, that the range of suavissima does not areas of secondary contact (Haffer 1970, 1974; extend much further into Brazil than currently Parsons et al. 1993) they have been considered documented, and that the lack of records of ei- by various authors to be a single biological spe- ther form in Amazonia west of Manaus reflects cies (Haffer 1970, 1974; Snow 1975, 1979) or a the actual distribution of the two species. group of distinct biological species (American The relationship between species limits and Ornithologist’s Union 1983; Hilty and Brown conservation biology has been outlined else- 1986; Stiles and Skutch 1989; Ridgley and where (e.g., Eldredge 1992, Zink 1993). In gen- Gwynne 1989; Prum 1990b, 1992; Parsonset al. eral, the degreeof precision of diversity estimates 1993). There is little agreement on how much and consequent,taxon-oriented conservation ef- morphological integration or interbreeding is re- forts are limited by the precision ofthe taxonomy quired to constitute grounds for combining dif- employed. Efforts to survey and catalogue avian ferentiated taxa into a polytypic species.Com- diversity in the Neotropics using the currently paring other characteristics, the amount of recognized species limits will not recover the syringeal differentiation between serena and contribution to patterns of endemism made by suavissimais more extensive than among other highly differentiated taxa that are combined in manakin congeners (Prum 1992). The conse- wide-ranging polytypic species,such as the for- quent differencesin vocalizations are extensive, merly polytypic Lepidothrix serena. There are and at least as divergent as other biological spe- dozens of other legitimate examples of differ- 700 RICHARD 0. PRUM entiated allopatric taxa that are currently rec- thologists would identify and record well-differ- ognized as subspeciesin the Neotropical subos- entiated taxa that are currently recognized as tines alone (e.g., holochlora litae, subspeciesif these forms were given speciessta- leucorrhoa altera, Laniisoma elegans tus, or if subspecificidentifications were actively buckleyi, Machaeropteru regulus striolatus, Py- encouraged. roderusscutatus occidentalis, Rupicola peruviana Conservation efforts require that omithologi- sanguinolenta; sensu Snow 1979). cal publications encourage the use of the most Unfortunately, subspeciesare viewed by many specificidentification possiblegiven the methods ornithologists as an inconsequential category of of observation or investigation. Although many diversity, or as too narrowly differentiated to be subspecies are difficult to identify even with identified without specimens,special experience, voucher specimens, ornithological publications or extra effort. Use of subspecific names is ac- should encourage, rather than discourage, their tively discouragedby many ornithological insti- use. These identifications should be accompa- tutions. Contributors to this and other major nied by documentation of how the determination American ornithological journals are instructed, was made. Additional researchon specieslimits “Do not give subspecificidentification unless it in polytypic Neotropical taxa is also essential to is pertinent and has been critically determined” furthering our understanding of patterns of en- (Anon. 1992). Since ornithologists frequently do demism. Although the needs of conservation bi- not view subspeciesas pertinent to their research ology cannot be considered a primary justifica- and generally believe that “critical” determina- tion for recognizing phylogenetic or more finely tion requires voucher specimens that are infre- differentiated biological species,it should be rec- quently collected, much of what we could know ognized that maintaining the taxonomic status about Neotropical birds has been obscured. quo could constitute a significant cost to conser- As an example, several publications in the last vation efforts in the Neotropics. two decadeshave documented new localities of the polytypic White-fronted Manakin, Lepidoth- ACKNOWLEDGMENTS rix serena, in Brazil, near Manaus (Willis 1977, I would like to thank G. F. Barrowclough (AMNH) Willis and Oniki 1988, Stotz and Bierregaard and R. L. Zusi (USNM) for loans of specimens used 1989, Bierregaard 1990). These records repre- in the research.Greg Budney of the Cornell Laboratory of Natural Sounds provided field recordingsof vocal- sented a range extension of over 500 km from izations. Doug Wechsler of Visual Resourcesfor Or- the nearest previously known locality (Blake nithology provided photographic evidence. Helpful 1950). In none of the publications was the sub- comments on the manuscript were made by Richard species of serena established. None mentioned C. Banks, Mark B. Robbins, and Douglas F. Stotz. Funds for the researchwere provided by the Graduate whether voucher specimens of serena had been ResearchFund of the University of Kansas. collected or deposited in any museum. To de- termine whether nominate serena or suavissima LITERATURE CITED extended its range from the Guianas to the Ma- AMERICANORNITHOLOGISTS UNION.’ 1983. Check- naus region, I had to rely on photographs de- list of North American Birds, 6th Ed. Washington, posited at Visual Resources for Ornithology DC. (VIREO, Academy of Natural Sciencesof Phil- ANONUMOUS. 1992. Information for contributors. Condor 94:311-312. adelphia), and undocumented memories of re- BIERREGAARD,R. O., JR. 1990. Speciescomposition searcherswho had been to the area. and trophic organization of the understory This lack of appropriate documentation of avi- community in a central Amazonian terra firme an diversity in the Neotropics has resulted in part forest, p. 217-236. In A. H. Gentry [ed.], Four from reliance by professional ornithologists on Neotropical rainforests. Yale Univ. Press, New Haven, CT. inadequate, current speciestaxa. Current species BLAKE,E. R. 1950. Birds of the Acary Mountains of names are insufficient to characterize many of southern British Guiana. Fieldiana lZool.1 32(7): _ . ~I the units of diversity in the Neotropical avifauna, 417494. and they constitute a significant burden to the CAPPARELLA,A. P. 1988. Genetic variation in Neo- documentation and analysis of avian biogeo- tropical birds: implications for the speciationpro- cess,p. 1658-l 664. In H. Ouellet [ed.], Acta iXX graphic patterns that are critical to establishing Congr. Int. Omithol. Nat. Mus. Nat. Sci., Ottawa. diversity estimates and conservation priorities. CAPPARELLA,A. P. 1991. Neotropical avian diversity In many cases, amateur and professional omi- and riverine barriers, p. 307-3 16. In B. D. Bell SPECIES STATUS OF THE WHITE-FRONTED MANAKIN 701

[ed.], Acta XX Congr. Int. Omithol. New Zealand McKrratc~, M. C., AND R. M. ZINK. 1988. Species Omithol. Congr. Trust, Christchurch. conceptsin ornithology. Condor 90:1-14. CHUBB.C. 1921. The birds of British Guiana. Vol. MEYERDE SCHAUENSEE, R.--l 966. The speciesof birds IL’Quaritch, London. of South America with their distribution. Livina- CRACK, J. 1983. Speciesconcepts and speciation ston Press, Narbeth, PA. analysis,p. 159-l 87. In R. F. Johnston [ed.], Cur- MEYER DE SCHAUENSEE,R., AND W. H. PHELPS,JR. rent ornithology, vol. 1. Plenum Press,New York. 1978. Guide to the birds of Venezuela. Princeton CRACRAFT,J. 198% Biological diversification and its Univ. Press, Princeton, NJ. causes.Ann. Missouri Bot. Gard. 721794-822. MOSKO~ITS,D., J. W. FITZPATRICK,AND D. E. WIL- ELDREDGE,N. [ED.] 1992. Systematics,biogeography, LARD. 1985. Lista preliminar das avesda Estacao and the biodiversity crisis. Columbia Univ. Press, Ecologicade Maraca. Pap. Avulsos Zool.(Sao Pau- New York. lo) 36:5 l-68. FROST,D. R., AND D. M. HILLIS. 1990. Species in NELSON,G. J., ANDN. I. PLATNICK. 1981. Systematics concept and practice: herpetologicalapplications. and bioaeoaranhv:cladistics and vicariance. Co- Herpetologica 46:87-104: _ _ _ lumbia Un&. Press, New York. HACKETT,S. J. 1993. Phylogeneticand biogeographic NOVAFS.F. C. 1978. Omitoloaia do Territ6rio do relationships in the Neotropical genus Gymnopi- Amapa. II. Publ. Avulsas Mu;. Para. Emilio Goel- thys (Formicariidae). Wilson Bull. 105:301-3 15. di No. 29. HACKETT. S. J.. AND K. V. ROSENBERG.1990. Evo- PARSONS, T. J., S. L. OLSON,AND M. J. BRALJN.1993. lution of South American antwrens (Formicari- Unidirectional spread of secondarysexual plum- idae): comparisonsof phenotypic and genetic dif- agetraits acrossan avian hybrid zone. Science260: ferentiation. Auk 107:473-489. 1643-1645. HAFFER,J. 1970. Art-Entstehungbei einigen Wald- PRUM,R. 0. 1985. Observationson the White-front- vijgeln Amazoniens. J. Omithol. 111:285-33 1. ed Manakin (Pipra serena)in Suriname. Auk 102: HAFFER.J. 1974. Avian speciationin SouthAmerica. 384-387. P&l. Nuttall Omithol. Club. No. 14. Cambridge, PRUM, R. 0. 1988. Historical relationships among MA. avian forest areasof endemism in the Neotropics, HAVERSCHMIDT,F. 1968. The birds of Surinam. Ol- p. 2562-2572. In H. Ouellet [ed.], Acta IXX Congr. iver and Bovd. Edinburah. U.K. Int. Omithol. Nat. Mus. Nat. Sci., Ottawa. HELLMAYR, C. E: 1906. A ievision of the speciesof PRUM, R. 0. 1990a. A test of the monophyly of the the genusPipra. Ibis 1906:l-46. manakins (Pipridae) and (Cotingidae) HELLMA~R,C. E. 19 10. Fam. Pipridae, p. l-3 1. In basedon morphology.Oct. Pap. Mm. Zool., Univ. P. Wytsman [ed.], Genera avium, vol. 9. V. Ver- Mich. 722. teneuil and A. Desmet, Brussels. Paul, R. 0. 1990b. Phylogeneticanalysis of the evo- HELLMAYR,C. E. 1929. Catalogue of birds of the lution of display behavior in the neotropical man- Americas, part 6. Oxyruncidae-Pipridae-Cotingi- akins (Aves: Pipridae). Ethology 84:202-23 1. dae-Rupicolidae-Phytotomidae. Publ. Field Mus. FRUM,’ R. 0. 1992. Syringealmorphology, phylogeny, Nat. Hist. 266. and evolution of the neotropical manakins (Aves: HILLY, S. L., AND W. L. BROWN. 1986. A guide to Pipridae). Amer. Mus. Nat. Hist. Novitates 3043. the birds of Colombia. Princeton Univ. Press, RIDGELY,R. S., AND J. A. GWYNNE. 1989. A guide Princeton, NJ. to the birds of Panama. Princeton Univ. Press, JOHNSTON,N. K. 1980. Charactervariation and evo- Princeton, NJ. lution of sibling speciesin the Empidonax d@- hX+WAY, R. 1907. The birds of North and Middle cilis-flavescenscomplex. Univ. of California Publ. America. Part IV. Bulletin United StatesNat. Mus. Zool. 112:1-151. 50. KROODSMA,D. E. 1981. Learning and the ontogeny ROBBINS,M. B., ANDR. S. RIDGELY. 1992. Taxonomy ofsound sianalsin birds.__ D. l-23. In D. E. Kroods- and natural history of Nyctiphrynus rosenbergi ma and E. H. Miller [eds.], Acoustic communi- (Caprimulgidae). Condor 94:984-987. cation in birds, vol. 2. Academic Press,New York. SALOON,O., ANDF. D. GODMAN. 1882. Notes on birds KROODSMA,D. E. 1984. Songsofthe Alder Flycatcher from British Guiana. Ibis 1882:76-84. (Empidonax alnorum) and Willow Flycatcher SM~THE,F. B. 1975. Naturalist’s color guide. Amer. (Empidonax trailliz] are innate. Auk 101: 13-24. Mus. Nat. Hist., New York. LANYON,W. E. 1978. Revision of the Myiarchus fly- SNOW,D. W. 1975. The classificationof the mana- catchersof South America. Bull Amer. Mus. Nat. kins. Bull. Brit. Omithol. Club 95:20-27. Hist. 161:427-628. SNOW,D. W. 1979. Tityrinae, Pipridae, Cotingidae, LINNEAUS. 1766. SystemaNaturae. 12th ed. Uppsala, p. 229-308. In M. A. Traylor, Jr. [ed.], Check-list Sweden. of birds of the world, vol. 8. Mus. Comp. Zool., MAYR, E. 1942. Systematicsand the origin of species. Harvard Univ., Cambridge, MA. Columbia Univ. Press, New York. STEPHENS,L., AND M. A. TRAYLOR,JR. 1985. Omi- MAYR. E. 1969. Princiulesof systematiczoologv. _- Mc- thological gazetteer of the Guianas. Mus. Comp. Graw Hill, New York. _ Zool.; Harvard Univ., Cambridge. MA. MAYR, E., AND W. H. PHELPS,JR. 1967. The origin STILESF. G.. AND A. F. SKUTCH. 1989. A auide to of the bird fauna of the south Venezuelan high- the birds of Costa Rica. Princeton Univ. Press, lands. Bull. Amer. Mus. Nat. Hist. 136:269-328. Princeton, NJ. 102 RICHARD 0. PRUM

STOTZ, D. G., ANDR. 0. BIERREGAARD,JR. 1989. The WILLIS,E. 0. 1977. Lista preliminar das aves da pane birds of Fazendas Port0 Alegre, Esteio, and Di- noroestee Areasvizinhas da Reserva Ducke, Ama- mona, north of Manaus, Amazonas, Brazil. Rev. zonas, Brazil. Rev. Brasil. Biol. 37:585-601. Brasil. Biol. 49:861-872. WILLIS,E. O., ANDY. ONIKI. 1988. Aves observadas STOTZ,D. G. 1993. A hybrid manakin (Pipra) from em Balbina, Amazonas e OSprovlveis efietos da Roraima, Brazil, and a phylogenetic perspective barragem. Ciencia e Cultura 40:280-284. on hybridization in the Pipridae. Wilson Bull. 105: ZINK, R. M. (CONVENOR)1993. Speciesconcepts. Spe- 348-351. cial Symposium, 11lth Annual Meeting, Amer. WILEY,E. 0. 1978. The evolutionary speciesconcept Omithol. Union, Fairbanks, AK. reconsidered.Syst. Zool. 27: 17-26.