Tyrannida II

Passerines

Tyranni: Suboscines

Passeri: Oscines

Passerida

Sylvioidea
Muscicapoidea and allies
Passeroidea

The 44 Orders

Paleognaths

Galloanserae

Metaves

Pelecanae

Charadriae

Passerae

Tyrannidae: Tyrant Flycatchers

105 genera, 411 species HBW-9

The Tyrannidae are the largest of the suboscine families, including about one third of all of the suboscines. It is also one of the most difficult to classify, full of cryptic species. Some flycatchers are so cryptic that there remains controversy about whether 1 or 2 species are involved, such as the the Cordilleran and Pacific-slope Flycatchers.

Fitzpatrick presents a traditional taxonomy of the Tyrannidae in HBW (2004a), augmented by a new tribe, Contopini. This taxonomy has considerable truth in it, although it ignores completely a major clade (Platyrinchinae) and oversimplifies the true complexity of flycatcher relationships.

The rebuilding of Tyrannidae starts by separating the Tityridae in their own family, then moves on to include the deep split between the Platyrinchinae (aka Pipromorphinae) and the rest, the Tyranninae. Broader and deeper genetic studies have uncovered additional complexity. For one, the split between Platyrinchinae and the rest may not be the basal division in the Tyrannidae. Rather, one of our troublesome groups comes first. Ericson et al. (2006b) and Ohlson et al. (2008) found that Piprites is sister to the rest of the Tyrannidae, which can then be divided into the Platyrinchinae and a clade containing Tyranninae. Although we follow that here, Tello et al. (2009) found a different result, with Piprites relatively basal within Platyrinchinae.

Tyrannidae
Click for genus-level tree
for Tyrannidae

A Platyrinchus/Neopipo clade is basal within the Platyrinchinae. Although Tello et al. (2009) consider Neopipo unrelated to Platyrinchus, Rheindt et al. (2008a) and Ohlson et al. (2008) found evidence they are sister taxa. All three analyses used different genes, so I'm going with the majority here. There is also a question about whether this group is closer to the other Platyrinchinae or not as one of the genes analyzed by Rheindt et al. (2008a) placed it outside this clade and closer to the remaining Tyrannidae. The other gene analyzed by Rheindt et al. agrees with the placement I use, as do Ericson et al. (2006b), Ohlson et al. (2008) and Tello et. al. (2009), so I have pretty high confidence that it belongs there.

The placement of Platyrinchus affects the nomenclature. It means the clade is called Platyrinchinae (probably Bonaparte 1854, perhaps Olph-Gaillard 1857), rather than Rhynchocyclidae (Berlepsch, 1907). Neither Mionectinae (Sibley & Ahlquist 1985), or Pipromorphinae (Wolters 1977) due to priority. Pipromorphinae is often attributed to Bonaparte 1853 or or 1854, giving it priority. This is not valid on the technical grounds that the genus Pipromorpha (Gray 1855) had not yet been named. If the term had been in common use, this wouldn't matter much, but it seems to have remained completely unused until revived by Wolters (1977). This unusual circumstance means that the first valid use of this old term was in 1977! Its priority dates to that.

The striking Many-colored Rush Tyrant, Tachuris rubrigastra, although sometimes considered and Elaeniid, seems to be sister to the rest of Platyrinchinae. The remainder of the Platyrinchinae can be divided into two clades: the Mionectine flycatchers (Pipromorphini, which Fitzpatrick hides in the unrelated Elaeniini) and a tody-tyrant/flatbill/twistwing clade.

Genetic analysis shows three groups in the Pipromorphini. Three sets of genes have been used to investigate this, by Ohlson et al. (2008), Rheindt et al. (2008a), and Tello et al. (2009). Unfortunately, they give three different results for which group is basal! I don't see any reason to choose between them, and treat them as an unresolved trichotomy. Note that I follow Miller et al. (2008) and re-split Mionectes into Pipromorpha and Mionectes. I also divide Pogonotriccus Bristle-Tyrants from Phylloscartes, as is commonly done (but not by SACC, yet).

The tody-tyrant/flatbill/twistwing clade has two main pieces, Todirostrini and Rhynchocyclini, and the twistwing group (Cnipodectes and Taeniotriccus), provisionally called Cnipodectini. Rheindt et al. (2008a), Ohlson et al. (2008), Tello and Bates (2007), and Tello et al. (2009) again give different arrangements of the three clades. Tello et al. (2009) is alone in placing Cnipodectini away from the Todirostrini and Rhynchocyclini. It seems best to put them in a trichotomy for now.

There are some changes within the tody-tyrants (Todirostrini). The whole Hemitriccus group (including Lophotriccus, Myiornis, and Oncostoma) is rearranged based on Tello and Bates (2007), and Rheindt et al. (2008a). Lophotriccus (including the former Atalotriccus) is merged into Oncostoma, as are Hemitriccus minor and spodiops. Five species are retained in Hemitriccus, which is sister to the expanded Oncostoma. The Boat-billed Tody-Tyrant moves to the genus Microcochlearius, which is sister to Myiornis. The remainder of Hemitriccus goes back into a restored Idioptilon, which is sister to the Microcochlearius/Myiornis clade. The changes in species names moving to Oncostoma and Idioptilon are due to the fact that both are neuter in gender. There's some guesswork here as to which species go where, so expect further changes once more study is done of these species.

Next on the list is Hirundineinae. Ohlson et al. (2008) considered this group closest to the Tyranninae and Fluvicolinae. In contrast, Tello et al. (2009) found them sister to the Elaeniinae. Finally, Rheindt et al.'s (2008a) data put them in a more basal position, sister to the combined Elaeniinae, Tyranninae, and Fluvicolinae. Until this is clarified, they must remain in their own subfamily. Given the lack of consensus, it makes the most sense to put them in a trichomoty with Elaeniinae and Tyranninae/Fluvicolinae. Ohlson et al. considered all six species, so I follow their arrangement of the species. Note that the three former Myiophobus included in this subfamily are not closely related to the rest of Myiophobus. They have been given a new genus name, Nephelomyias, by Ohlsson et al. (2009).

Elaeniinae has two major branches: Euscarthmini and Elaeniini. Ohlson et al. (2008) and Tello et al. (2009) find very similar results for the Euscarthmini. Rheindt et al. (2008a) is a bit different, but not very strongly so. Sixteen species of Phyllomyias tyrannulets are separated from the main Phyllomyias group (part of Elaeniini) and placed variously in the revived genera Tyranniscus and Xanthomyias. The “Mercocerculus” in Euscarthmini are not closely related to the Mercocerculus in Elaeniini. The arrangement of species within Zimmerius is based on Rheindt et al. (2008c), although they did not sample cinereicapilla and villarejoi. They also found evidence of a yet undescribed species hidden in Z. chrysops.

The situation in the Elaeniini is rather complicated, with different analyses giving quite different results. I don't see good reasons to choose between them, so I'm treating it as a polytomy with five clades, ordered by size. The clades are: (1) Suiriri; (2) the Nesotriccus-Phyllomyias group; (3) Tyrannulus and the Myiopagis elaenias; (4) the Elaenia elaenias; and (5) the Pseudelaenia-Serpophaga group. Ohlson et al. (2008) group (1)-(4) together, Tello et al. (2009) group (1), (3), (4) together sister to (2) plus (5), while Rheindt et al. (2008a) group (1), (2) and (5), with (3) sister to them, and (4) sister to the rest. Finally, although their taxon sampling was quite sparse, Ericson et al. (2006b) have (3) and (5) closer to each other than to (4).

The Monte Tyrannulet / Gray-crowned Tyrannulet, Serpophaga griseicapilla, has had a somewhat checkered history. It was previously referred to as Serpophaga griseiceps, but the type of griseiceps was actually a juvenile of S. munda (Herzog and Mazur Barnett, 2004). Straneck (2007) gave it the name S. griseiceps. It has been referred to as Gray-crowned Tyrannulet, but SACC proposal #419 prefers Monte Tyrannulet, which is also in use.

The ordering within Elaenia itself is based on Rheindt et al (2008b, 2009a), which also indicates that the Roraiman Elaenia should be split from the Sierran Elaenia, and that the subspecies sorida should be split from the Highland Elaenia E. obscura. This subspecies, which currently lacks an English name, would be sister to the Great Elaenia, E. dayi. At present, SACC does not recognize it and it is not yet included in the list. The Myiopagis Elaenias are organized as in Rheindt et al. (2009b)

This brings us to the clade including Fluvicolinae and Tyranninae. The first question concerns the position of the Short-tailed Field Tyrant, Muscigralla brevicauda. It has long been a controversial species. Some have even suggested it is an ant-thrush, while others have argued it should be merged into Muscisaxicola (for a discussion, see Lanyon, 1986). DNA analyses are not immune to this controversy. Ohlson et al. (2008) located it basally in Fluvicolinae, and Tello et al. (2009) used a different set of genes to locate it basal to both Fluvicolinae and Tyranninae. At the other extreme, Chesser (2000), using a third set of genes, found it sister to Muscisaxicola, and there was evidence they should be merged! A compromise here is surely wrong. The taxonomy here follows Chesser, but keep in mind that it could be completely wrong.

The attilas are basal in Tyranninae, and there is a kingbird group, and a Myiarchus group. After that, the picture is a little less clear as there are differences between Ohlson et al. (2008) and Tello et al. (2009). The latter has better taxon sampling in this group, so the tree I'm using is closer to Tello et al. However, I've kept Legatus close to the kingbird group, and treated Ramphotrigon (including Deltarhynchus) as an independent group. Oddly, Ohlson et al. put Ramphotrigon sister to the Fluvicolinae, while Rheindt et al. (2008a) and Tello and Bates (2007) put it with the kingbirds or Myiarchus group, respectively. All this uncertainty leads to a separate clade for Ramphotrigon.

Once that is done, the rest of the Tyranninae are fairly straightforward, and there is at least agreement over what genera are in each clade. I've put the Lesser Kiskadee in Philohydor as it is not certain that Pitangus is its closest relative (e.g., Chaves et al., 2008). I've also separated the Fork-tailed Flycatcher in Milvulus based on Tello et al. (2009). Once it is out of Tyrannus it reclaims its old name and becomes Milvulus tyrannus, not Milvulus savana. The arrangement of Myiarchus is based on Joseph et al. (2004). They note some issues within Myiarchus. There are also some species boundaries that need to be redrawn. One such case involves La Sagra's and Stolid Flycatchers. They seem to be mutually paraphyletic and either need to be further split or regrouped (Joseph et al. 2004). Other such problems exist for ferox, phaeocephalus, swainsoni, and tuberculifer.

There's substantial difference between Ohlson et al. (2008) and Tello et al. (2009) concerning the Fluvicolini. In particular, Tello et al. nest Ohlson et al.'s clade F-2 within their clade F-1. Further, the position of the troublesome taxa Colonia and Sublegatus differs. Rheindt et al. (2008a) also considered Colonia and Sublegatus, but did not include any other fluvicoline flycatchers. However, they weren't near anything else or each other, so their results suggest the two are either in Fluvicolini or more basal, but not in Contopini or Tyranninae. Ohlson et al. have Sublegatus sister to Contopini/Xolmini and Colonia in one of the Fluvicolini clades (F-2). In contrast, Tello et al. put Colonia as a basal clade in Fluvicolinae while Sublegatus was sister to the F-2 clade.

Obviously, it is uncertain where Colonia and Sublegatus go, and there are issues with how the other pieces of the Fluvicolini fit together (if they fit together at all). I'm following a compromise topology which maintains the F-1 and F-2 clades of Ohlson et al. (2008), but treating them as sisters with F-2 augmented by Sublegatus. Colonia is placed outside the Fluvicolini, as in Tello et al. (2009). It's probably wrong, but I don't see any good options here.

In the F-1 clade, Silvicultrix is separated from Ochthoeca (see García-Moreno and Arctander, 1998) and several of the former Myiophobus are temporarily designated “Myiophobus”. It's possible that Tumbezia should be merged into Ochthoeca, but I'd prefer more information first.

The Contopini are fairly straightforward, with Cicero and Johnson (2002), Ohlson et al. (2008) and Tello et al. (2009) telling the same basic story. The only real disagreement concerns whether Sayornis and Mitrephanes are sister taxa. Note also that the true Myiophobus are basal in Contipini. Finally, Empidonax taxonomy follows Cicero and Johnson (2002) and Johnson and Cicero (2002).

That brings us to the last tribe, Xolmini. The order here follows Tello et al. (2009), except for the insertion of Muscigralla as mentioned above. The order in Ohlson et al. (2008) is a bit different, with the biggest difference the position of Lessonia.

Pipritinae: Piprites

Platyrhynchinae: Mionectine Flycatchers [Pipromorphinae]

Platyrhynchini: Spadebills

Tachurini: Many-colored Rush Tyrant

Pipromorphini

Cnipodectini: Twistwings

Rhynchocyclini: Flatbills

Todirostrini: Tody-Flycatchers

Hirundineinae

Elaeniinae: Elaenine Flycatchers

Euscarthmini: Tyrannulets

Elaeniini: Elaenias, Tit-Tyrants

Tyranninae: Tyrant Flycatchers

Fluvicolinae: Fluvicoline Flycatchers

Coloniini: Long-tailed Tyrant

Fluvicolini: Fluvicoline Flycatchers

Contopini: Pewees, Phoebes, Empids

Xolmiini: Ground and Bush Tyrants

Previous Page Next Page