Proposal (810) to South American Classification Committee

Transfer Rynchops into Laridae and treat as a subfamily (Rynchopinae)

Background: Whether to place Rynchops in Laridae or in its own family has been historically controversial and intermingled with the familial status of terns. SACC maintains a classification in which terns are merged with Laridae but Rynchops is maintained in its own family Rynchopidae.

New information: Molecular studies have found Rynchops to be sister to Laridae (Paton et al. 2003) or Sternidae (Paton & Baker 2006, Fain & Houde 2007, Prum et al. 2015) but more recently Sternidae has been found not to be monophyletic because Anous and Gygis are outside a clade that includes Laridae, Sternidae, and Rynchops (Baker et al. 2007, Odeen et al. 2010). Only one recent analysis found Rynchops outside a Sternidae-Laridae clade but only in analyses using mitochondrial genes (Jackson et al. 2012). Further, the analysis was somewhat simplistic (e.g. no partition by codon position), taxon sampling was poor (10 species), and they didn’t include Anous. Using 259 loci, Prum et al. (2015) found Rynchops sister to Sterna, with Larus more basal.

Recommendation: The precise affinities of Rynchops are not well resolved. but maintaining Rynchops in a separate family makes Laridae paraphyletic in most scenarios. Therefore, I recommend placing Rynchops into Laridae as a subfamily Rynchopinae.

References:

Baker, A. J., S. L. Pereira & T. A. Paton 2007. Phylogenetic relationships and divergence times of Charadriiformes genera: multigene evidence for the Cretaceous origin of at least 14 clades of shorebirds. Biology Letters 3: 205-210.

Fain, M. G., & P. Houde 2007 Multilocus perspectives on the monophyly and phylogeny of the order Charadriiformes (Aves). BMC Evol. Biol. 7: 35.

Jackson, D. G., S. D. Emslie, & M. Van Tuinen. 2012. Genome skimming identifies polymorphism in tern populations and species. BMC Research Notes 5(1): 94.

Ödeen, A., O. Håstad & P. Alström 2009. Evolution of ultraviolet vision in shorebirds (Charadriiformes). Biology Letters 6(3): 370-374.

Paton, T. A., A. J. Baker, J. G. Groth & G. F. Barrowclough 2003. RAG-1 sequences resolve phylogenetic relationships within Charadriiform birds. Mol. Phylogenet. Evol. 29: 268-278.

Paton, T. A., & A. J. Baker 2006 Sequences from 14 mitochondrial genes provide a well-supported phylogeny of the Charadriiform birds congruent with the nuclear RAG-1 tree. Mol. Phylogenet. Evol. 39: 657-667.

Prum, R. O., J. S. Berv, A. Dornburg, D. J. Field, J. P. Townsend, E. Moriarty Lemmon & Alan R. Lemmon. 2015 A comprehensive phylogeny of birds (Aves) using targeted next-generation DNA sequencing. Nature 526(7574): 569.

Santiago Claramunt, January 2019

Comments from Areta: “Perhaps. But what are we supposed to do with Anous and Gygis? We need to decide this in a comparative framework to achieve consistency. We could either give family ranks to Anous, Gygis and Rynchops or consider them as subfamilies of Laridae as done by Cracraft (2013). Making the move of Rynchops without deciding what to do with the noddies and the White Tern seems problematic.”

Comments from Stotz: “NO. My sense of this is the evidence so far suggests that we are wrong to split out Rynchopidae as a separate family, but the appropriate treatment remains obscure. Nacho’s comment about noddies and White Tern are on the money, given that the basis for bringing Rynchops into Laridae is that they appear to be outside of the Larinae-Sterninae clade. If that is the case, then it makes no sense to hold onto Rynchopinae as a subfamily unless we are willing to erect subfamilies for Anous and Gygis.”

Additional comments from Claramunt: “The non-monophyly of the subfamily Sterninae is a separate problem that could be deal with in a follow-up proposal. Of course this proposal does not resolve that problem because this proposal is focused on the non-monophyly of Laridae due to the position of Rynchops. Once we decide on Rynchops, we could quickly move to deal with Gygis and Anous.

Comments from Robbins: “YES. We could go either way with this. Yes, Nacho and Doug make good points, but I agree with Santiago in dealing with Rynchops now, then follow up with another proposal for Gygis and Anous. So, I vote for transferring Rynchops into Laridae.”

Comments from Remsen: “NO. The important point that Santiago makes is that there is no evidence to maintain Rynchops is the sister to all members of Laridae and that as long as we include Gygis and Anous in Laridae, we have a paraphyletic Laridae. The problem, however, may be generated entirely by Gygis and Anous, which if removed from Laridae might make Rynchopidae a valid family level taxon if sister to all other Laridae with thorough taxon- and gene-sampling. Santiago’s assessment is that the burden-of-proof now falls on the stance of maintaining Rynchopidae as a separate family, and I agree with this in principle. Nonetheless, given the unexpected finding that Gygis and Anous are not terns (they are indeed weirdos with respect to other terns), I prefer a slow-it-down, more conservative approach until we have at least one additional study that includes all relevant taxa and a time-calibrated phylogeny. Prum et al. (2015) estimated the split between gulls and the ancestor of Sterna and Rynchops as mid-Miocene, with a subsequent (very short branch length) split of Sterna and Rynchops as a few million years later; Gygis and Anous were not sampled (and only one genus of tern and gull). Before we make a decision, I personally would be more comfortable with an analysis that includes all of these taxa as well as at least a few additional genera of gulls and terns. Then, we would have all the information available for making a better-informed decision. At this point, assuming the initial findings that Gygis and Anous are the oldest lineages in this entire group, the branching pattern might be so tight that a single expanded Laridae for all of them, including Gygis and Anous, might be the best treatment.

“Hopefully, our public discussion of this will be sufficient provocation for someone to undertake this study.”

Comments from Zimmer: “NO. I think we need more information (taxon and gene sampling) to resolve the position of Gygis and Anous relative to Larinae, Sterninae, and Rynchops before we start messing with Rynchops. It does appear, as the proposal (and Doug) suggest, that our current treatment of Rynchops is incorrect, but better in my opinion, to not compound that error by changing it up, only to have to reverse course when we get better resolution regarding the noddies and White Tern. I think that this could be one of those cases where, if we find ourselves in a hole with a shovel, the best thing to do is stop digging.”

Comments from Cadena: “YES to place Rynchops in Laridae, but NO to recognize Rynchopinae as a subfamily. Like I have insisted many times before but seems not to lead us to discuss the issue, I think we need to agree on how exactly are we going to recognize clades above the species level, with subfamilies being an especially thorny case as I described most recently in comments under proposal 778 and a few earlier ones. I think that unless we recognize subfamilies consistently across all families following clear criteria, there is very little value –and lots of grounds for disagreement for entirely subjective reasons– in engaging in this sort of issues. This case is not quite the same as others because I realize that we already recognize subfamilies in Laridae so the issue would come up of where do we place Rynchops, but then we would have to work to revise the classification of subfamilies to make sure they are monophyletic because Sterninae does not appear to be so.”

“Given the discussions about slowness of SACC etc. over the last few weeks, my personal take is that we should focus on the really pressing issues where changes in classification are truly required. I think we all agree in that we should recognize families, correct? But do we need to recognize subfamilies? Will users of our classification who are eager to see us move faster will benefit from having subfamilies? If the committee decides we do need to get into this out of principle or because it would be useful for the community, then I think we need to work on it across all clades and do it consistently, i.e. a lot of work. I personally think we have our plate pretty full with genera, families, and orders (let alone all the problems with species limits!), so I would rather not deal with subfamilies at all given that there is no objective definition of what a subfamily is and especially given that our classification does not consistently recognize subfamilies within all (or most) families. I am not saying we do know what families (or orders or genera) are, but at least we have consistently recognized clades within orders as such across the board.

“Maybe I am the only one who thinks that a discussion on whether and how we recognize higher-level clades (and well, species, but that’s for another day) is necessary, so this will be my final rant on this and will give up if people don't think it matters.”

Comments from Jaramillo: “NO. This group is one of the most unusual on earth based on its bill shape. It does not look right as a “mere” subfamily. One may argue that this creates a scenario that is equally troubling as other gull/tern like birds would end up being outside of the gulls and terns (Gygis and Anous), and that seems wrong as well. But looking further out in the group, jaegers (Stercorarius) are also rather gull like, but clearly outside of the gulls, as they are sister to alcids. At a distance young jaegers are mistaken for gulls other than by experienced observers. So perhaps having other gull/tern like birds outside of the gulls and terns is not all that much of an issue, the gull/tern type is recurrent in the group. We deal with that later perhaps by creating a new family for Gygis and Anous? They are not very tern like vocally in my opinion.”