OSME Region List of birds

White-throated Robin Irania gutturalis © Aurélien Audevard

Ornithological basis for the list

Version 3.3α

The compilation of a species list for any major region requires its authors to address questions of taxonomy, and the OSME Region List (ORL) is no exception. Taxonomic decisions above the level of species are best made in a global context; hence the ORL originally was based on Dickinson (2003) Howard & Moore 3rd edn for Taxonomic Order and Sequence, and on Gill & Wright (2006) Birds of the World: Recommended English Names – the published list of the International Ornithological Congress [IOC] regional sub-committees. We review the frequent IOC on-line updates (at, v6.3 at time of writing) that have supplanted Gill & Wright (2006). The updates summarise well-referenced taxonomic, order, sequence and name changes, both accepted and proposed. We adopt these revisions with few exceptions, and document our reasoning. We have reviewed the ORL having evaluated Howard & Moore 4th edn Vols 1 & 2 at length. Many conclusions and decisions in H&M4 require assessment by IOC, but their extent and complexity will take much more time had been anticipated, and so we will follow IOC’s graduated approach, given that IOC have their own priorities. That said, the differences are mostly of degree, and are not radical. Furthermore, H&M4 conclusions will need evaluating in light of relevant papers published after H&M4 cut-off dates. We believe our approach not only presents the most up-to-date and well-researched methods of world lists available to us but also is accessible to birders in general. One of the ORL Team, Mike Blair, was recently invited on to the IOC World Bird List Advisors’ Panel.

The main focus to our taxonomic work for the ORL has therefore been around species-level taxonomy. We have adopted a framework similar to that adopted by the British Ornithologists’ Union’s Taxonomic Subcommittee, whereby taxa are classed either as species, allospecies or semispecies within a superspecies, or subspecies within a species (see below). For the vast majority of taxa, allocation of taxonomic rank was straightforward. For these taxa, binomial and trinomial nomenclature is used for species and subspecies respectively; we have indicated allospecies and semispecies by inclusion of the name of the superspecies to which they belong in square brackets between the genus and species name (see fuller explanation below). We departed from Dickinson (2003) on relatively few occasions, mostly to incorporate peer-reviewed taxonomic changes, cited in the ORL ‘Notes’ column. In similar fashion, occasionally we have departed from the IOC World Bird List, but we have included their recommended English names in curly brackets {…}.

For some taxa, however, the decision is not clear-cut, and here we faced a number of choices. Indeed, much recent research has indicated that many more taxa than previously thought are in this category; eg see Olsson et al (2010) (as analysed by Bannikova 2010) on the large grey shrikes. They highlight the possible danger of relying on a single molecular marker, eg mtDNA, in taxonomic revisions and phylogenetic inferences, stating, “Since the mitochondrial gene tree deviates substantially from the (non-cladistic) interpretation of relationships based on morphological and ecological characteristics, and there are indications that the gene tree might not fully conform with the organismal phylogeny, any proposed taxonomy is uncertain”. However, a number of recent authoritative references (up to 2014) have not diverged from our earlier interpretation.

To add to the difficulties, an increasing number of taxa have been shown to have two lines of ancestry, inferring secondary contact between two long-separated populations when barriers to interbreeding have not developed, eg Sternkopf et al 2010.

We could have decided to ‘force’ these taxa into one rank or another (by choosing a default rank to assign in such cases, by following the majority position of previous authors, by tossing a coin in each case, or by other equally arbitrary methods), but we felt that that was an unscientific approach. At the other extreme, we could have chosen to study each case in detail and come up with our own considered view based on the available evidence. This avoids unnecessary delays in publishing each version of the ORL.

We have therefore deferred a number of decisions until such time as further information is available. In such cases, we feel that our pragmatic approach clearly identifies this treatment as interim, and that we might be able to make more definite decisions in future. However, it is clear that sometimes the evolutionary state of some taxa is such that they are essentially undefinable in terms of rank, but we are strongly of the opinion that they should continue to be presented separately because they represent coherent populations, often over a wide geographical area.

In support of our approach, we would cite van Deemter (2010), who elegantly demonstrates the circumstances in which the use of conceptual vagueness provides better understanding than a fruitless search for a ‘one-size-fits-all’ precise definition, citing taxa relationships as the prime example; for most taxa, rank is a ‘convenient fiction’ (van Deemter 2010) that causes little problem.

Determination of the status of the taxa listed

Our approach in the ORL presents practical ways of presenting the status of the taxa listed. It is not a definitive taxonomy in any way, but seeks not only to identify where knowledge of any taxon is imperfect, but also to keep it from being overlooked. The status of each taxon was considered to fall into one of the following categories:

  1. Full biological speciesie those taxa which are (to all intents and purposes) completely reproductively isolated – eg Sardinian Warbler Sylvia melanocephala and Rüppell's Warbler S. ruppeli.
  2. Subspeciesie taxa which have separate geographic ranges, look different from each other in some way, where we want to recognise this variation, but where all are undoubtedly just different forms of the same species – eg two subspecies of Common Buzzard Buteo buteo are buteo and vulpinus (Steppe Buzzard) where trinomial names are used without need for any brackets – eg Buteo buteo vulpinus.
  3. Allospeciesie taxa that have separate geographic breeding ranges (often in the past being classed as subspecies of a single species). These are taxa where we have ‘good reasons’ to believe that we are dealing with taxa, which if they were sympatric (ie sharing part of their geographic ranges), would behave as full biological species – eg there are differences in appearance, vocalisations or behaviour (in any combination) and in habitat or genetics (or both), which differences, if taken together, are comparable to those between known full species. Because such taxa are allopatric, we are making informed judgments, but in treating two taxa as allospecies, we are making a confident statement that we believe the evidence is good enough to warrant this status, which we identify by using square brackets […] between the genus and species names. Allospecies come in pairs or groups, the group being called a superspecies.
  4. Superspecies – a group of allospecies or semispecies. As for allospecies, we put the superspecies name in square brackets […] between the genus and species names – eg Asian Desert Warbler Sylvia [nana] nana and African Desert Warbler Sylvia [nana] deserti.
  5. Semispecies – these are like allospecies, but come into contact in a hybrid zone. An extensive hybrid zone, spread over a large geographic area, would be a strong indicator that we are dealing with subspecies. What marks out semispecies as different, and warranting treatment as "fully tickable", is that there is something within the hybridisation acting as an effective barrier in keeping the two taxa apart (eg hybrids are less fit). Like allospecies, semispecies come in pairs or groups, and the term superspecies again applies to the group. The same convention applies as for allospecies – square brackets [...] – eg Carrion Crow Corvus [corone] corone and Hooded Crow Corvus [corone] cornix. With semispecies, we are again making a confident statement that we believe we have enough evidence for that judgement.
  6. Ring species – a ring species is a species that consists of a series of neighbouring populations, where no population is reproductively isolated from its neighbours, except for the two ‘end’ populations in the series, which are sympatric and reproductively isolated, and thus appear to be separate species – the name derives from the geographical distribution of the populations, which usually forms a ring.. A well-known example is that of the Greenish Warbler Phylloscopus [trochiloides] trochiloides where the five subspecies – viridanus*, ludlowi*, trochiloides, obscuratus and plumbeitarsus* (those asterisked are recorded from the OSME Region) form a series around the Himalayas from northwest to northeast – with the northwesternmost (viridanus) occurring sympatrically with the northeasternmost (plumbeitarsus) without interbreeding. The square brackets indicate that Greenish Warbler forms a superspecies with Green Warbler P.[t.] nitidus.
  7. "Don't know" – our shorthand description of the final category. This is where the taxa could be full species, allospecies/semispecies, subspecies, or exist in identifiable populations that are intermediate in evolutionary terms. Also, it may be that we haven’t formed an opinion, because the evidence is not available, contradictory, or we are unsure whether it has been shown to apply sufficiently comprehensively (We may not have had time to discuss which treatment is appropriate). Here we use round brackets (…) for the "Don't knows". In effect we are saying that – eg we use round brackets for Pied Wagtail Motacilla (alba) yarrelli, because there are a number of possible approaches, each of which may be correct, but we don't yet know which one, and so the end result may be Motacilla alba as a full species, Motacilla alba yarrelli as a subspecies, or Motacilla [alba] yarrelli as an allospecies or semispecies.

Improved understanding of the relationships between species causes inevitable changes to checklist sequences and will continue to do so, some genera moving to a different family, or families themselves being subsumed in others. All world checklists are subject to such revisions – these arise from morphology, vocalisation studies, molecular biochemistry and other disciplines. Popular understanding of the limitations of DNA research is often poor, partly because the results are not easy to interpret and concern probabilities. Morphological differences do not invariably coincide geographically with DNA ‘breaks’ that show good separation ‘distances’ between taxa, although so far this is rare. We seek corroboration from other disciplines where use of a single DNA technique offers radical conclusions.

We would expect DNA barcode analysis (Kerr et al 2007) of Palearctic species will provide some unexpected conclusions (Kevin Kerr pers comm) over the coming decade.

The use of scientific names from the ORL

Square-bracketed scientific names

As explained above, we use square brackets […] in the ORL between the genus and species names in the scientific names of allospecies or semispecies, and in superspecies, which are groups of such allospecies and semispecies. However, there is a general convention of omitting the square-bracketed part of the scientific name to simplify them in general correspondence and in papers that do not deal with taxonomic aspects. You can see this in the output of the most respected ornithological organisations, such as the British Ornithologists’ Union (BOU), for example, in the Fifth Report of the Taxonomic Sub-committee of the BOU Records Committee, where the split is accepted of Dark-throated Thrush Turdus ruficollis into Black-throated Thrush T. atrogularis and Red-throated Thrush T. ruficollis. In the ORL, we present this information in a slightly different way, as semispecies: we add a Parent Taxon row, to show more clearly the former treatment from which the change derives, thus: Dark-throated Thrush Turdus ruficollis, and then the semispecies rows thus: Black-throated Thrush Turdus [ruficollis] atrogularis and Red-throated Thrush Turdus [ruficollis] ruficollis.

There is no expressed intention that users of the ORL should include the square-bracketed part of a taxon’s scientific name except when discussing it with us.

Round-bracketed scientific names

Our use of round brackets (…) between the genus and species names in scientific names in the ORL indicates that there is a sizeable element of uncertainty as to the taxonomic status of the form in question. We would suggest users of the ORL follow suit in correspondence and in scientific papers. We invite comment on our taxonomic decisions (or perhaps more important, on the non-decisions) to help us improve the ORL. In particular, we are keen to hear views on those taxa which are confined or largely confined to the OSME Region. Whilst a considerable amount of work has been done on the taxonomy of birds in the northern Palearctic, and much is ongoing, less attention has been focussed on taxonomic problems relevant to the OSME Region. We hope that the OSME Region List can provide a vehicle through which this can be addressed.