Proposal (754) to South American Classification Committee:

 

Elevate 13 taxa to species rank based on playback experiments

 

A.  Elevate Pseudocolaptes johnsoni to species rank

B.  Elevate Automolus virgatus to species rank

C.  Elevate Grallaria alticola to species rank

D.  Elevate Scytalopus intermedius to species rank

E.   Elevate Ochthoeca thoracica to species rank

F.   Elevate Myadestes venezuelensis to species rank

G.  Elevate Pheugopedius schulenbergi to species rank

H.  Elevate Amazonian populations of Tunchiornis ochraceiceps to species rank

I.     Elevate South American populations of Basileuterus culicivorus to species rank

J.   Elevate Myiothlypis chlorophrys to species rank

K.  Elevate Myiothlypis striaticeps to species rank

L.   Elevate Atlapetes tricolor crassus to species rank

M. Elevate Amazonian populations of Arremon aurantiirostris to species rank

 

 

A prelude to a set of 13 proposals: background and methods from Freeman & Montgomery (2017):

 

Background:

Geographically isolated populations of birds often differ in song. Because birds often choose mates based on their song, song differentiation between isolated populations constitutes a behavioral barrier to reproduction. If this barrier is judged to be sufficiently strong, then isolated populations with divergent songs may merit classification as distinct species under the biological species concept.

 

Methods:

We conducted playback experiments to measure whether populations discriminated against song from a related, allopatric population - these experiments that simulated secondary contact between geographically isolated populations. Briefly, each experiment measured the behavioral response of a territorial bird to two treatments: 1) song from the local population (sympatric treatment) and 2) song from the allopatric population (allopatric treatment). All territorial birds responded to sympatric song by approaching the speaker (typically to within 5 m).

 

We defined song discrimination as instances in which the territory owner(s) ignored allopatric song, defined as a failure to approach within 15 m of the speaker in response to the allopatric treatment. We calculated song discrimination for each taxon pair as the percentage of territories that failed to approach the speaker in response to allopatric song. For example, a song discrimination score of 0.8 indicates that 80% of territorial birds (e.g. 8 out of 10) ignored allopatric song while simultaneously actively defending a territory. We assume that song discrimination is a proxy for premating reproductive isolation; that is, our experiments provide insight into whether these populations would recognize each other as conspecific and interbreed (or not) were they to come into contact with one another. It is unknown what degree of song discrimination is “enough” that song constitutes a strong enough premating barrier to reproduction that allopatric populations merit classification as distinct biological species. To provide a yardstick, we considered nine allopatric Neotropical taxon pairs that were recently split (or have pending proposals to SACC) in part based off differences in vocalizations. We found the average song discrimination in these nine taxon pairs to be ~ 0.6 (60% of territorial birds ignored allopatric song), and suggest that species limits deserve to be reconsidered when taxon pairs currently classified as subspecies have song discrimination scores above ~ 0.6.

 

In most cases, we played songs of populations A and B (where populations A and B comprise a taxon pair) to territorial birds of population A. That is, in the majority of cases we asked whether population A discriminated against song from population B but not the reverse. To date, we have conducted reciprocal playback experiments for 23 taxon pairs (13 oscines and 10 suboscines, see Table S2) in which we measured both discrimination of population A to song from population B and also discrimination of population B to song from population A and in which we conducted at least four experiments on each population (n = 11.5 Ī 3.8 playback experiments/population, range = 4 – 23, see Table S2). Song discrimination in these reciprocal cases was highly correlated (r = 0.88, t = 8.4, df = 21, p-value < 0.0001, Figure 1). There were no examples with strongly asymmetric song discrimination (e.g. population A discriminates against population B, but population B does not discriminate against population A). We therefore assume that unidirectional data accurately describes song discrimination within taxon pairs in our database.

 

 

Please forgive the brevity of the following proposals. I did not have time to give each proposal the detail it deserves; nonetheless, I believe the “bare-bones” approach of these proposals provides sufficient relevant information to consider re-evaluating species limits in the following 14 cases.

 

Benjamin Freeman, 15 September 20

 

 

 

 

 

754A. Elevate Pseudocolaptes johnsoni to species rank

 

Effect on SACC: Elevate Pseudocolaptes johnsoni to species rank (split from Central American Pseudocolaptes lawrencii).

 

Background: Related populations of Pseudocolaptes (tuftedcheeks) occur in Central America (lawrencii) and western Colombia and Ecuador (johnsoni). These allopatric populations are currently classified as conspecific, but differ in plumage and voice. Due to these differences in mate choice traits, they have sometimes been considered to represent two distinct species (e.g., HBW).

 

New data: Freeman and Montgomery (2017) conducted playback experiments on 10 territories of Central American lawrencii. They found that 9 out of 10 territorial birds discriminated against song playback of South American johnsoni (discrimination = fail to approach within 15 m of the speaker. For canopy species such as Pseudocolaptes, we calculated distance to speaker in the horizontal plane. That is, a bird perched high in the canopy but directly above the speaker would have a distance to speaker of 0 m). Perhaps relevant is that these two populations differ in ND2 sequences (mtDNA) deposited on GenBank by 8.5% in uncorrected p-distance, which suggests these two populations may have evolved strong postmating isolation as well, although this is speculative.

 

Competing proposals:

1)   NO = Retain the status quo.

2)   YES = Elevate Pseudocolaptes johnsoni to species rank (split from Central American Pseudocolaptes lawrencii).

 

Recommendation: Central American birds respond strongly to lawrencii song, but essentially ignore song from johnsoni. This suggests that vocal differences constitute a strong premating barrier to reproduction between these taxa. I therefore recommend treating johnsoni and lawrencii as distinct biological species. English names currently in use are Pacific Tuftedcheek for johnsoni and Buffy Tuftedcheek for lawrencii. However, at present “Buffy Tuftedcheek” refers to johnsoni + lawrencii. I am agnostic on English names for this (and every other) proposal.

 

 

 

754B. Elevate Automolus virgatus to species rank

 

Effect on SACC: Elevate Automolus virgatus to species rank (split from Amazonian Automolus subulatus).

 

Background: Related populations of Automolus woodhaunters occur west of the Andes (virgatus group in Central America and western Colombia and Ecuador), and east of the Andes (Amazon basin; subulatus group). These allopatric populations are currently classified as conspecific by SACC, but differ markedly in voice, and somewhat in plumage. Due to these differences, they have often been considered to represent two distinct species (e.g., HBW, many field guides).

 

New data: Freeman and Montgomery (2017) conducted playback experiments on 12 territories of virgatus group individuals (6 in Costa Rica and 6 in western Ecuador). They found that 11 out of 12 territorial birds discriminated against song playback of the Amazonian subulatus group (discrimination = fail to approach within 15 m of the speaker).

 

Competing proposals:

1)   NO = Retain the status quo.

2)   YES = Split Automolus virgatus from A. subulatus.

 

Recommendation: Central American and Choco birds respond strongly to local song, but essentially ignore song from Amazonian birds. This suggests that vocal differences constitute a strong premating barrier to reproduction between these taxa. I therefore recommend treating the virgatus group and the subulatus group as distinct biological species. English names currently in use are Western Woodhaunter for the virgatus group, and Eastern Woodhaunter and Amazonian Woodhaunter for the subulatus group.

 

 

 

754C. Elevate Grallaria alticola to species rank

 

Effect on SACC: Elevate Grallaria alticola to species rank (split from G. quitensis).

 

Background: Grallaria quitensis inhabits high elevations (typically paramos) in the Northern and Central Andes. Three allopatric subspecies differ in vocalizations, and, to a lesser extent, plumage.

 

New data: Freeman and Montgomery (2017) conducted playback experiments on 9 territories of quitensis in Ecuador to measure response to alticola (restricted to the Eastern Andes in Colombia). They found that 6 out of 9 territorial birds in Ecuador discriminated against song playback of alticola (discrimination = fail to approach within 15 m of the speaker). Perhaps relevant is that these two populations differ in ND2 sequences deposited on GenBank by 5.4% in uncorrected p-distance, which suggests that the two populations may have evolved reasonably strong postmating isolation as well, though this is speculative.

 

Competing proposals:

1)   NO = Retain the status quo.

2)   YES = Elevate Grallaria alticola to species rank.

 

Recommendation: Birds from Ecuador respond strongly to local song, but, in the majority of cases, ignore song from andicola birds. This suggests that vocal differences constitute a strong premating barrier to reproduction between these two taxa. I therefore recommend treating alticola and quitensis as distinct biological species. I am not aware of existing English names; HBW uses “Northern Tawny Antpitta” for alticola and “Western Tawny Antpitta” for quitensis; there may be better options.

 

 

 

 

754D. Elevate Scytalopus intermedius to species rank

 

Effect on SACC: Elevate Scytalopus intermedius to species rank (split from S. latrans).

 

Background: Scytalopus latrans inhabits high elevations in the Northern and Central Andes. Song is somewhat divergent between multiple populations. This proposal focuses on two groups; the population of latrans found in southeastern Ecuador, and intermedius, which occurs in Peru south of the MaraĖon Gap.

 

New data: Freeman and Montgomery (2017) conducted playback experiments on 8 territories of latrans in southeastern Ecuador. They found that all 8 territorial birds in Ecuador discriminated against song playback of intermedius (discrimination= fail to approach within 15 m of the speaker).

 

Competing proposals:

1)   NO= Retain the status quo.

2)   YES = Elevate Scytalopus intermedius to species rank.

 

Recommendation: Birds from southeastern Ecuador respond strongly to local song, but, in all cases, ignore song from intermedius. This suggests that vocal differences constitute a strong premating barrier to reproduction between these two taxa, which are found on adjacent and opposite sides of the MaraĖon Gap. I therefore recommend treating populations north and south of the MaraĖon Gap as distinct biological species. There is marked variation in vocalizations within latrans found north of the MaraĖon Gap, some of which may be relevant in the future to species limits within the latrans group (= all population found north of the MaraĖon Gap). I am not aware of existing English name.

 

 

 

 

754E. Elevate Ochthoeca thoracica to species rank

 

Effect on SACC: Elevate Ochthoeca thoracica to species rank (split from O. cinnamomeiventris).

 

Background: Ochthoeca cinnamomeiventris is widely distributed in the tropical Andes. Vocal and plumage variation is divergent between multiple populations. This proposal focuses on two taxa; cinnamomeiventris, found north of the MaraĖon Gap, and thoracica, which occurs in Peru south of the MaraĖon Gap. These taxa differ in song and plumage.

 

New data: Freeman and Montgomery (2017) conducted playback experiments on 5 territories of cinnamomeiventris in southeastern Ecuador. They found that all 5 territorial birds in Ecuador discriminated against song playback of thoracica (discrimination= fail to approach within 15 m of the speaker).

 

Competing proposals:

1)   NO = Retain the status quo.

2)   YES = Elevate Ochthoeca thoracica to species rank.

 

Recommendation: Birds from southeastern Ecuador respond strongly to local song, but, in all cases, ignore song from thoracica. This suggests that vocal differences constitute a strong premating barrier to reproduction between these two taxa, which are found on adjacent and opposite sides of the MaraĖon Gap. I therefore recommend treating cinnamomeiventris and thoracica as distinct biological species. An English name currently used for thoracica is “Maroon-belted Chat-Tyrant”; as far as I can tell, cinnamomeiventris is known as “Slaty-backed Chat-Tyrant,” which is also the English name widely used for the entire complex.

 

 

 

754F. Elevate Myadestes venezuelensis to species rank

 

Effect on SACC: Elevate Myadestes venezuelensis to species rank (split from M. ralloides).

 

Background: Myadestes ralloides is widely distributed in the tropical Andes. Vocal variation is markedly divergent among populations of M. ralloides. This proposal focuses on two taxa; venezuelensis, found north of the MaraĖon Gap in eastern Ecuador, and ralloides, which occurs in Peru south of the MaraĖon Gap.

 

New data: Freeman and Montgomery (2017) conducted playback experiments on 12 territories of venezuelensis in eastern Ecuador. They found that 10 out of 12 territorial birds in Ecuador discriminated against song playback of ralloides (discrimination= fail to approach within 15 m of the speaker).

 

Competing proposals:

1)   NO = Retain the status quo.

2)   YES = Elevate Myadestes venezuelensis to species rank.

 

Recommendation: Birds from eastern Ecuador respond strongly to local song, but, in nearly all cases, ignore song from ralloides. This suggests that vocal differences constitute a strong premating barrier to reproduction between these two taxa, which are found on adjacent and opposite sides of the MaraĖon Gap. I therefore recommend treating populations north and south of the MaraĖon Gap as distinct biological species. I am not aware of existing English names. There is further vocal variation between populations living on the Pacific slope in western Ecuador and those on the Amazonian slope in eastern Ecuador – these may also prove to constitute distinct biological species, but our (unpublished) playback experiments between these taxa is not sufficient to evaluate species limits at present.

 

 

 

754G. Elevate Pheugopedius schulenbergi to species rank

 

Effect on SACC: Elevate Pheugopedius schulenbergi to species rank (split from P. euophrys).

 

Background: Pheugopedius euophrys is widely distributed in the Northern Andes. Vocal and plumage variation is somewhat divergent between multiple populations. This proposal focuses on two taxa; longipes, found north of the MaraĖon Gap in eastern Ecuador, and schulenbergi, which occurs in Peru south of the MaraĖon Gap.

 

New data: Freeman and Montgomery (2017) conducted playback experiments on 19 territories of longipes in southeastern Ecuador. They found that 12 out of 19 territorial birds in Ecuador discriminated against song playback of schulenbergi (discrimination= fail to approach within 15 m of the speaker).

 

Competing proposals:

1)   NO = Retain the status quo.

2)   YES = Elevate Pheugopedius schulenbergi to species rank.

 

Recommendation: Birds from southeastern Ecuador respond strongly to local song, but, in most cases, ignore song from schulenbergi. This suggests that vocal differences constitute a strong premating barrier to reproduction between these two taxa, which are found on adjacent and opposite sides of the MaraĖon Gap. I therefore recommend treating populations north and south of the MaraĖon Gap as distinct biological species. Schulenbergi is called “Grey-browed Wren” by HBW.

 

 

 

754H. Elevate Amazonian populations of Tunchiornis ochraceiceps to species rank

 

Effect on SACC: Elevate Amazonian populations of Tunchiornis ochraceiceps to species rank (split from populations found west of Andes).

 

Background: Tunchiornis ochraceiceps consists of many named subspecies that are found both west and east of the Andes. There is marked vocal and plumage variation throughout its broad distribution. 

 

New data: Freeman and Montgomery (2017) conducted playback experiments on 10 territories of T. ochraceiceps in Costa Rica. They found that 8 out of 10 territorial birds discriminated against song playback from the western Amazonian ferrugineifrons group (discrimination = fail to approach within 15 m of the speaker).

 

Competing proposals:

1)   NO = Retain the status quo.

2)   YES = Split populations found west of the Andes from populations found east of the Andes.

 

Recommendation: Central American birds respond strongly to local song, but essentially ignore song from western Amazonian birds. This suggests that vocal differences constitute a strong premating barrier to reproduction between these allopatric taxa. It would be nice to have done further playback experiments in western Colombia and western Ecuador, but this was not possible. There are additional vocal differences within both west-of-the-Andes and east-of-the-Andes groups. In sum, though this is not as clear-cut as many of the other cases, I recommend treating the populations found east and west of the Andes as distinct biological species. I am not aware of existing English names that would be applied to west-of-the-Andes vs. east-of-the-Andes groups.

 

 

 

754I. Elevate South American populations of Basileuterus culicivorus to species rank

 

Effect on SACC: Elevate South American populations of Basileuterus culicivorus to species rank (split from populations found in Central America).

 

Background: Basileuterus culicivorus consists of many named subspecies found throughout much of Central and South America. There is marked vocal and plumage variation throughout its large distribution. This proposal focuses on two allopatric taxa; godmani, found in Costa Rica and Panama, and the cabanisi group which occurs in nearby Colombia and Venezuela.

 

New data: Freeman and Montgomery (2017) conducted playback experiments on 17 territories of godmani in Costa Rica. They found that 13 out of 17 territorial birds discriminated against song playback from the cabanisi group (discrimination = fail to approach within 15 m of the speaker).

 

Competing proposals:

1)   Retain the status quo.

2)   Elevate Basileuterus cabanisi to species rank.

 

Recommendation: Central American birds respond strongly to local song, but essentially ignore song from cabanisi. This suggests that vocal differences constitute a strong premating barrier to reproduction between these allopatric taxa. Thus, I recommend treating the populations found in South America as distinct from those in Central America. I note that species limits within South American populations may require further evaluation. English names: HBW uses “Stripe-crowned Warbler” for Central American populations, “Yellow-crowned Warbler” for populations in Colombia and Venezuela, and “Golden-crowned Warbler” for the additional South American populations. We have no playback data to evaluate differences between the cabanisi group and other populations in South America at this time. Still, the data we summarize above suggest that Central American populations represent a distinct species from those in adjacent northwestern South America; thus this proposal.

 

 

 

754J. Elevate Myiothlypis chlorophrys to species rank

 

Effect on SACC: Elevate Myiothlypis chlorophrys to species rank (split from M. chrysogaster).

 

Background: Myiothlypis chrysogaster consists of two populations, one found in the Choco in western Colombia and western Ecuador, and a second that inhabits eastern Peru (and, narrowly, Bolivia).  These taxa are notably different in vocalizations and have been considered to be distinct biological species by various authors.

 

New data: Freeman and Montgomery (2017) conducted playback experiments on 11 territories of chlorophrys in western Ecuador. They found that 10 out of 11 territorial birds discriminated against playback of chrysogaster song (discrimination = fail to approach within 15 m of the speaker). We also recently conducted two playback experiments that showed that chrysogaster discriminates against (ignores) chlorophrys song (unpublished, and small sample size caveat, but I thought worth mentioning).

 

Competing proposals:

1)   NO = Retain the status quo.

2)   YES = Elevate Myiothlypis chlorophrys to species rank.

 

Recommendation: Chlorophrys responds strongly to local song, but essentially ignores song from chrysogaster. This suggests that vocal differences constitute a strong premating barrier to reproduction between these allopatric taxa. Thus, I recommend treating these two populations as distinct biological species.  These taxa are sometimes referred to by the English names “Choco Warbler” (for chlorophrys) and “Cuzco Warbler” (for chrysogaster).

 

 

 

754K. Elevate Myiothlypis striaticeps to species rank

 

Effect on SACC: Elevate Myiothlypis striaticeps to species rank (split from M. luteoviridis).

 

Background: Myiothlypis luteoviridis is widely distributed in the tropical Andes. Vocal variation is somewhat divergent between multiple populations. This proposal focuses on two taxa; luteoviridis, found north of the MaraĖon Gap in eastern Ecuador, and striaticeps, which occurs in Peru south of the MaraĖon Gap.

 

New data: Freeman and Montgomery (2017) conducted playback experiments on 11 territories of luteoviridis in southeastern Ecuador. They found that 8 out of 11 territorial birds in Ecuador discriminated against song playback of striaticeps (discrimination= fail to approach within 15 m of the speaker).

 

Competing proposals:

1)   NO = Retain the status quo.

2)   YES = Elevate Myiothlypis striaticeps to species rank.

 

Recommendation: Birds from southeastern Ecuador respond strongly to local song, but, in most cases, ignore song from striaticeps. This suggests that vocal differences constitute a strong premating barrier to reproduction between these two taxa, which are found on adjacent and opposite sides of the MaraĖon Gap. I therefore recommend treating populations north and south of the MaraĖon Gap as distinct biological species. I am not aware of existing English names.

 

 

 

754L. Elevate Atlapetes tricolor crassus to species rank

 

Effect on SACC: Elevate Atlapetes tricolor crassus to species rank 

 

Background: Atlapetes tricolor consists of two populations, one found in the Choco in western Colombia and western Ecuador, and a second that inhabits eastern Peru (and, narrowly, Bolivia).  These taxa are notably different in vocalizations (and morphology) and have been considered distinct biological species by several authors.

 

New data: Freeman and Montgomery (2017) conducted playback experiments on 6 territories of crassus in western Ecuador. They found that 5 out of 6 territorial birds discriminated against playback of tricolor song (discrimination = fail to approach within 15 m of the speaker).

 

Competing proposals:

1)   NO= Retain the status quo.

2)   YES = Elevate Atlapetes crassus to species rank.

 

Recommendation: Atlapetes crassus responds strongly to local song, but essentially ignores song from tricolor. This suggests that vocal differences constitute a strong premating barrier to reproduction between these allopatric taxa. Thus, I recommend treating these two populations as distinct biological species. “Choco Brush-Finch” has been used for crassus and “Tricolored Brush-Finch” for tricolor.

 

 

 

754M. Elevate Amazonian populations of Arremon aurantiirostris to species rank

 

Effect on SACC: Elevate Amazonian populations of Arremon aurantiirostris to species rank (split from populations found west of Andes).

 

Background: Arremon aurantiirostris consists of many named subspecies that are found both west and east of the Andes. There is marked vocal variation throughout its distribution.

 

New data: Freeman and Montgomery (2017) conducted playback experiments on 23 territories of occidentalis in western Ecuador, and playback experiments on 12 territories of spectabilis in eastern Ecuador. They found that 16 out of 23 territories of occidentalis discriminated against playback of spectabilis song, and that 8 out of 12 territories of spectabilis discriminated against playback of occidentalis song  (discrimination = fail to approach within 15 m of the speaker).

 

Competing proposals:

1)   NO = Retain the status quo.

2)   YES = Elevate Amazonian Arremon aurantiirostris spectabilis to species rank.

 

Recommendation: Arremon aurantiirostris populations in western and eastern Ecuador respond strongly to local song, but, in most cases, ignore song from each other. This suggests that vocal differences constitute a strong premating barrier to reproduction between occidentalis and spectabilis. Thus, I recommend treating these two populations as distinct biological species. I note that this complex likely harbors additional biological species (e.g., in Central America, populations on the Caribbean and Pacific slope show strong song discrimination). Thus, if this split is adopted, it should be recognized that this is not the end of the story (including choosing English names that would minimize confusion in the case of additional splits).

 

 

 

Comments from Stiles: "This is a most interesting proposal!  However, while highly suggestive of biological isolating mechanisms, in most of the individual cases presented I consider that the data presented are insufficient for SACC acceptance, for the following reasons:

 

1) Most of the cases presented involved presenting the song of one population to members of another; reciprocal playbacks would have been more convincing, as asymmetry in responses have been found in some other studies;

 

2) The 60% no-response threshold presented was insufficiently justified: just how much variation exists in this threshold (which was based upon a general average of nine such comparisons of subspecies vs. species responses); it would be helpful to know how much variation occurred around this value, and whether the “threshold” might vary between oscines vs. suboscines, especially given that several of the comparisons presented barely exceed this “threshold”.

 

3) Several of the comparisons were made between populations on either side of the MaraĖón gap; it is often not clear how close these were to the “dividing line” (in at least one case, a subspecies closer to the gap exists, but was not sampled). In other cases, the author did not specify how closely the respective populations approach each other, and how distant the sampled populations were from the point of closest approach;

 

4) No spectrographic analyses were made of multiple recordings to evaluate possible intrapopulation variation, which might influence the results in several cases; song dialects mediated by learning and cultural transmission are known in oscines;

 

5) In several cases, important distributional gaps occur between the populations sampled in which song variation is known to exist between sampled and unsampled populations (subspecies), which could easily influence interpretation of the results;

 

6) For nearly all cases, genetic data would have helped to clinch the decision, but only for two cases were such data presented;

 

7) The lack of nomenclatural precision often makes the results difficult to apply, as in a number of cases, the taxa involved are polytypic, which at a minimum would require a detailed study of distributions and name priorities essential for arriving at concrete names for the putative species; in several such cases, the subspecies sampled were not given;

 

8) For a number of cases, pertinent investigations are underway (or planned), especially where vocal differences of unsampled populations may very well exist (a similar case in another SACC proposal regarding Henicorhina leucosticta did not pass for this reason). I note that none of the experiments were performed in Colombia, which would have been a strategic area for deciding several cases. Such cases in the present proposal may best be taken as pointers for more comprehensive analyses.

 

"All this is not to say that any of the cases presented are incorrect in suggesting that two (or more!) biological species are involved, but the data are insufficient for SACC action as they stand.  It is also worth noting that several of these proposals were rejected by SACC due to the lack of detailed spectrographic analysis of vocalizations: the point at issue is whether playback experiments represent a sufficient surrogate or substitute for such analysis. I conclude that if the experiments show significant differences in responses to same vs. different populations, the results may be taken as valid. However, when the differences are very close to the threshold value or sample sizes are too small, I recommend caution in their acceptance. I now mention each case, with respect to the problems described above by letters:

 

754A: Pseudocolaptes-plumage, genetic and reciprocal playbacks all support this split, so YES.

 

754B: Automolus-data very suggestive; not clear which population of the subulatus group was taken (where in “Amazonia”?).  Were reciprocal playbacks done for the two virgatus populations, which are at nearly opposite ends of this group’s range? On the other hand, in proposal 40 treating this question, all members of the committee agreed that two species were indicated, given the vocal differences, but most preferred to await a study of sonograms (by Zimmer?) Although I am not aware that this study has been published, I consider that the experiments by Freeman constitute at least a partial surrogate that to me, tips the burden of proof onto those who would maintain these two as conspecific, hence YES.

 

754C: Grallaria-unfortunately, no experiments performed in Colombia, where the Eastern Andes population also differs vocally from that inhabiting the Central Andes, which would seem to be in limbo: the same as the SE Ecuador population or not? Therefore, the data here are incomplete; although genetic data suggest that the two populations sampled represent different species, data from the C Andes birds in Colombia are needed to complete the picture, and I suggest a tentative NO.

 

754D: Scytalopus-here again, the data represent just the tip of the iceberg, since both north and south of the MaraĖón gap there exist several subspecies with different plumages. Daniel Cadena has been collecting genetic and vocal data on as many named Scytalopus taxa as possible, and may be nearer to doing a comprehensive treatment of the latrans group. Hence, although the data show that at least two biological species may well exist in this assemblage, I feel that any decision should await more comprehensive data, so NO for now.

 

754E: Ochthoeca-another case where multiple, well-defined subspecies exist, especially N of the MaraĖón gap, according to plumage; presumably the Ecuador sample was of subspecies; a sample of only five experiments with thoracicus from S of the MaraĖón and here again, I consider the data very suggestive but incomplete: NO for now.

 

754F: Myadestes-nominate ralloides was named from Bolivia (from where were taken the recordings used for this form?); venezuelensis from Venezuela; SE Ecuador is a long way from its type locality. Here, a detailed examination of intraspecific variation in song from intermediate populations (e.g., in Colombia and, depending on the above, N Peru) of these oscines (where song learning and cultural transmission of song dialects could well occur) should be done, and genetic data might also help: once again, I consider the experiments presented here highly suggestive, but the data are as yet incomplete: NO for now.

 

754G: Pheugopedius-the SE Ecuador population (longipes, N of the MaraĖón gap) was tested against schulenbergi, S of this gap: however, another subspecies (atriceps) occurs closer to this gap than schulenbergi and was not sampled. Intrapopulation variation in songs was not mentioned. So, once more, the data are highly suggestive but gaps remain. Also, the observed non-response percentage (63%) was very close to the “threshold” (see above), so NO for now.

 

754H: Tunchiornis-a widespread species, the nominate subspecies found from Mexico to W Panama, whereas cis-Andean ferrugineifrons occurs in S Venezuela, Colombia and perhaps S to EC Peru. Although the exact site from which the sample recording(s?) of the latter were taken was not given, evidence for species status of cis- vs. trans-Andean populations sampled is certainly suggestive. However, given the wide range of both forms, I am uneasy because there could well be vocal variation within both forms; hence, I think that more populations should be sampled for a conclusive answer: NO for now.

 

754I: Basileuterus culicivorus-this is a highly polytypic species, with several named subspecies in both Middle and South America. The tested population was godmani of Costa Rica to W Panamá, but the provenance of the recording of the “cabanisi group” was not specified (at least three subspecies of culicivorus occur in Colombia and Venezuela, and several more in cis-Andean South America). Clearly, a much wider sample of the various subspecies would be desirable, along with an evaluation of within-subspecies variation in at least some of these. Population genomics within this species would also be very interesting. Hence, I consider the available evidence too incomplete to justify this split at present; NO for now.

 

754J: Basileuterus chrysothlypis- This experiment dealt with a northern, trans-Andean subspecies (chlorophrys) and a more southern cis-Andean subspecies (nominate chrysothlypis); these two are thus widely separated geographically, such that parapatry is not relevant (or possible).  No other subspecies exist in this group. Here, the vocal evidence is stronger (including a very small sample of reciprocal playbacks). Although sonograms and genetic evidence would surely clinch the case, the available data seem to me to tip the burden of proof towards conspecifity, hence a tentative YES.

 

754K:. Myiothlypis luteiventris-Another comparison of subspecies N vs. S of the MaraĖón gap. The song differences, as described by Schulenberg et al. (2009) certainly seem considerable, but with only 8 of 11 (73%) no-responses of the northern luteiventris to songs of the southern striaticeps, the results should be considered tentative at present. Also, a subspecies further south (euophrys) shows greater morphological differences from striaticeps as well as possibly greater differences in song. So: NO for now, pending more data.

 

754L:  Atlapetes-This case is similar to no. 10 in comparing cis- and trans-Andean populations, but with a much smaller sample size (6); given that Atlapetes songs tend to be complex, intraspecific variation could exist. I consider the available evidence to be very suggestive but not conclusive, so NO for now.

 

754M: Arremon-This is another tip-of-the iceberg case, because aurantiirostris includes multiple subspecies with subtle morphological differences between them over a wide range (including two from Costa Rica, that differ noticeably in song). Hence, this result represents a small part of a much bigger problem. Because Jorge AvendaĖo is now investigating this problem in depth, I believe that any action by SACC should await the results of this study: NO for now.

 

Comments from Cadena: “I am going to vote NO on all these proposals out of principle. I think that the hard work done by Ben Freeman is exremely important and interesting, and is arguably the best evidence we have to evaluate species limits in these groups of birds. However, I agree with all the concerns voiced by Gary, especially with that related to the 60% threshold for discrimination. I am not sure exactly what this means, but say that 60% of the time, females of taxon A meeting males of taxon B discriminate against them. What happens the rest of the time? If on 40% of the encounters females of A do not discriminate against males of B, then inter-taxa matings would be extremely frequent and one would expect hybridization to be rampant, no? Maybe not, but then one should carefully consider other sources of evidence together with the playback data, which takes me to a general suggestion I have on all this. I think that the “bare-bones” approach taken by Ben leaves several open questions and suffers importantly from lack of detail. My suggestion is that all this should be placed in a paper (or papers) where the ranges of the taxa involved are properly described, variation in different characters is analyzed, and nuances of each case are discussed in detail to reach more informed decisions. There are several potential theses projects here for undergraduates with an interest in integrative taxonomy of birds in which Freeman et al. have already made important progress.”

 

Comments from Jaramillo: “A – YES mainly due to fact that this is a suboscine, and the genetic data.

B – YES

C – YES. The issue of other populations in Colombia which may also differ, or perhaps muddy this picture can be dealt with when those data appear.

D – YES. Scytalopus are so clear cut in response to playback, that I find that this may be enough, even though there may be other populations that should be sampled in the future. Or other molecular data will arrive in the future.

E – NO. Kind of on the fence on this one. But 5 experiments seems awful few.

F – YES.  Note also that Miller et al. 2007 published mtDNA data that also suggests a split north and south of the MaraĖon gap. Paper is available here. https://www.researchgate.net/profile/Matthew_Miller3/publication/228644579_Historical_biogeography_of_the_new_world_solitaires_Myadestes_spp/links/00b7d521e0aefa5c7c000000.pdf

G – NO. I would like to see more data here, 12 out of 19 is a lot of birds that did react positively. More data is necessary in this case.

H – NO. Given the huge variation in this species, I think a more widespread set of experiments, is necessary.

I -  NO, but not because I do not believe that there are multiple species here. But this one is one of those complex ones with possibly multiple species involved, furthermore B. hypoleucus may in fact be conspecific with one of these forms. So, it is complicated enough that I would rather base changes on the Vilaca & Santos (2010) paper. Copy available here:      https://s3.amazonaws.com/academia.edu.documents/46692562/Biogeographic_history_of_the_species_com20160621-14843-6idjqj.pdf?AWSAccessKeyId=AKIAIWOWYYGZ2Y53UL3A&Expires=1522809006&Signature=X%2FzDXcNowKX2AzmZ6ykEM8dpuNc%3D&response-content-disposition=inline%3B%20filename%3DBiogeographic_history_of_the_species_com.pdf

J – YES

K – YES, but tentative.

L – YES. Available molecular data corroborates. Sanchez-Gonzales et al. (2014) found that crassus was in a group with A. melanolaemus and A. forbesi. While Klicka et al. (2014) sampled A. tricolor tricolor and found it was in an entirely different clade with schistaceus, latinuchus etc. Sanchez-Gonzales et al. (2014) also confirm clearly defined differences in song of tricolor and crassus.

 

Klicka, J., Keith Barker, F., Burns, K. J., Lanyon, S. M., Lovette, I. J., Chaves, J. A. & Bryson, R. W., Jr (2014). A comprehensive multilocus assessment of sparrow (Aves: Passerellidae) relationships. Molecular Phylogenetics and Evolution, 77, 177–182.

 

M- NO. Playback experiments not quite conclusive here, and a very complex situation.