Proposal (312) to South American Classification Committee


Proposals to recognize the genera (A) Dixiphia and (B) Ceratopipra


Literature Review:


In the first phylogenetic analysis of the Pipridae based on a cladistic analysis of syringeal morphology, Prum (1992) identified a polyphyletic genus Pipra, including three lineages: the serena species group, the species pipra, and a clade including the erythrocephala and aureola species clades. The serena species group was distantly related to other Pipra, and the genus Heterocercus was more closely related to the erythrocephala + aureola clades than was pipra. 


To render the taxa monophyletic, Prum recommended that the serena group be placed in the available genus Lepidothrix Bonaparte 1854, and that pipra be placed in the genus Dixiphia Reichenbach 1850. Prum also placed the aureola clade in the subgenus Pipra (type aureola) and erythrocephala clade in the subgenus Ceratopipra Reichenbach (type cornuta) within Pipra. The Lepidothrix recommendation was rapidly adopted, but Dixiphia has only occasionally been used. Ceratopipra has likewise not been adopted.


Bostwick (2000) presented a phylogenetic analysis of behavioral characters that supported the placement of the three Machaeropterus species as the sister group to the erythrocephala clade.


Recently, Rego et al. (2007) present the first moderately well sampled molecular phylogeny of the Pipridae. They found molecular evidence that:


--- Heterocercus is more closely related to the erythrocephala and aureola clades than is the species pipra.


--- the three Machaeropterus species are sister group to the erythrocephala clade.


These findings are each congruent with some previous findings of Prum and Bostwick.


Taxonomic Recommendations:

To make the genera of manakins monophyletic, then several changes are necessary. One alternative is to place the species pipra, Heterocercus, Machaeropterus and the aureola and erythrocephala clades into an expanded genus Pipra. There seems to be no enthusiasm for this in the literature or the community. Alternatively, two new genera should be new recognized:


(1) the species pipra should be placed in the genus Dixiphia Reichenbach 1850 (type leucocilla Linnaeus = pipra Linnaeus). (See Prum 1992).


(2) the erythrocephala clade including cornuta, chloromeros, mentalis, rubrocapilla and erythrocephala should be moved to the genus Ceratopipra Spix (type Ceratopipra cornuta Spix).


The genera Heterocercus and Machaeropterus should remain unchanged in composition but change position in sequence, but Pipra should be restricted to include aureola, filicauda, and fasciicauda only.


These two recommendations can be considered separately by the committee: (A) Dixiphia, (B) Ceratopipra.



Bostwick, K. S. 2000. Display behaviors, mechanical sounds, and their implications for evolutionary relationships of the Club-winged Manakin (Machaeropterus deliciosus). Auk 117:465-478.

Prum, R. O. 1992. Syringeal morphology, phylogeny, and evolution of the Neotropical manakins (Aves: Pipridae). American Museum Novitates 3043:65 pp.

RÊGO, P. S., J. ARARIPE, M. L. V. MARCELIANO, I. SAMPAIO, AND H. SCHNEIDER. 2007. Phylogenetic analyses of the genera Pipra, Lepidothrix and Dixiphia (Pipridae, Passeriformes) using partial cytochrome b and 16S mtDNA genes. Zoologica Scripta 2007: 1-11. (let Remsen know if you need a pdf)


Rick Prum, Sept. 2007





Comments from Nores: "YES, a las dos sub-propuestas. Queda bien claro en los dos tipos de análisis (siringeal y molecular) que Dixiphia es algo distinto de Pipra y lo mismo pasa con el grupo bien definido de especies negras con cabeza roja o amarilla que forma como un casco. Como este grupo no puede ser agrupado con las verdaderas Pipra, tiene necesariamente que tener otro nombre, o sea Ceratopipra. La otra opción de poner pipra, Heterocercus, Machaeropterus, y los clades aureola y erythrocephala en un amplio género Pipra es mucho menos aceptable. ¿Lo único que me preocupa es que si este análisis es totalmente confiable? ¿Qué pasa si aparece otro análisis molecular (como en el caso de Drymophila de la propuesta 300) que dice lo contrario? Casi sería mejor esperar."


Comments from Cadena: "NO on both. The proposal by Rick makes two assertions regarding the study by Rego et al.: (1) that "Heterocercus is more closely related to the erythrocephala and aureola clades than is the species pipra", and (2) that "the three Machaeropterus species are sister group to the erythrocephala clade". In addition, it makes the following assertion regarding the study by Botswick: (3) "a phylogenetic analysis of behavioral characters that supported the placement of the three Machaeropterus species as the sister group to the erythrocephala clade".  I think that some comments on these three assertions are needed in order to clarify the situation.


"Regarding (1), there must be a mistake because Figure 2 in Rego et al. shows the species pipra forming a clade with erythrocephala (and rubrocapilla and Machaeropterus), so in contrast to what the proposal says, pipra is more closely allied to erythrocephala than is Heterocercus, which is outside the erythrocephala-rubrocapilla-Machaeropterus-pipra clade. On the other hand, it is true that Heterocercus appears to be more closely allied to the aureola clade than is the species pipra, but the support for this relationship is weak (0.54 posterior probability). In addition to the low support for this relationship, Prum's (1992) morphological phylogeny suggests a different arrangement, as it places the aureola and erythrocephala clades as sister to each other.


"Regarding (2), first, from Rego et al. the relationships between Machaeropterus and erythrocephala-rubrocapilla and between species pipra and the clade formed by these three taxa do not appear to be strongly supported (posterior probabilities of 0.91 and 0.93, respectively -- these are suggestive but not significant, so I would say that strictly speaking, the position of pipra and Machaeropterus in the molecular phylogeny is unresolved). Second, it is important to note that Rego et al. only sampled one of the species of Machaeropterus, not the three of them as indicated in the proposal. Nonetheless, because morphological data (Prum 1992) support the monophyly of Machaeropterus (Botswick's behavioral study also only sampled one species according to her Figure 8), this is really not a very important point. The real complicating factor is that the different analyses are not consistent: molecules place the erythrocephala clade as "successively" sister to Machaeropterus and pipra (albeit with low support), whereas morphology places the erythrocephala clade as sister to the aureola clade, with Machaeropterus and pipra outside this clade. This leads me to point three:


"(3) The proposal says that Botswick's analysis of behavioral characters placed Machaeropterus as sister to the erythrocephala clade. However, if I interpret her figure 8 correctly, what she found was that Machaeropterus is sister to a clade formed by the erythrocephala and aureola clades, which are recovered as sister, a finding that is consistent with morphological analyses.


"So, to summarize, I think that there is some discordance between the molecular and morphological phylogenies, and some of the relationships implied by the molecular phylogeny are not strongly supported. Taking a totally conservative position, one could argue that the discordances and weak molecular data are sufficient to preclude the need of making the proposed taxonomic changes. In particular, morphology and behavior suggest that the erythrocephala and aureola clades are sister groups (a hypothesis that cannot be rejected with statistical confidence by the molecular data), so there is no cogent reason to separate these in two genera. On the other hand, the position of pipra is supported by morphological evidence published back in 1992 but this was never incorporated in our classification for whatever reason, and the new molecular data provide only weak support for accepting Dixiphia as a distinct genus. In sum, I will tentatively vote NO on both proposals for now, but I would be more than happy to reconsider my vote based on further comments by Rick and other committee members."


Comments from Stiles: "NO to both, at least for now. I think Daniel's arguments are cogent and should be addressed. Like him, I am prepared to change my vote if and when stronger and more consistent support for recognition of Dixiphia etc. should be forthcoming."


Comments from Remsen: "NO to both, tentatively. Heterocercus is such a bizarre manakin that I will hold out for another genetic data set (another gene) to make sure we're looking at species' trees, not gene trees, that confirms that Heterocercus is closer to the erythrocephala and aureola clades than is P. pipra. Such a remarkable finding requires additional independent data, in my opinion. Daniel's points regarding Ceratopipra convince me to wait for additional data on this one as well."


Comments from Stotz: "NO to both. I am tempted by Dixiphia, but not really by Ceratopipra, but I think given some of the inconsistencies in the molecular vs. morphological data, I'd like to at least see some stronger molecular support for a particular topology. Like Van, I am a little surprised by the idea that Heterocercus is in the middle of what used to be "Pipra".


Comments from Pacheco: "NO to both. O contraponto apresentado por Cadena é demasiadamente convincente."