Birds of Madagascar (1 Viewer)

Hawkins, F., Safford, R. and Skerrett, A. (2015) Birds of Madagascar and the Indian Ocean Islands (Helm Field Guides)

New book awaiting publication has a small section entitled ‘New Genera in-waiting’ wherein the following spp are discussed:

‘Falco’ zoniventris Banded Kestrel (“Large Falco genus may deserve splitting into more genera”)

‘Sarothrura’ watersi Slender-billed ‘Flufftail’ ("Lemurolimnas")

‘Ninox’ superciliaris White-browed Owl (Authors say no genus name “has ever been proposed for this species” – however, Cephaloglaux Kaup, 1852 seems to apply)

‘Zoonavena’ grandidieri Madagascar Spinetail (“Presumed congeners [give] the genus a rather unlikely distribution; research should test for true relationship or convergence, and resulting possibility of different genera, among the three species”. NB: Z. grandidieri is the type species of Zoonavena so 'Zoonavena' would only apply to the aforementioned 'presumed congeners').

‘Mirafra’ hova Madagascar Lark (“recently shown to belong with…Eremopterix. Given extreme plumage distinctiveness…might merit its own genus”)

‘Ploceus’ nelicourvi Nelicourvi Weaver (“[Ploceus] probably requires subdivision… Nelicourvius was originally established for this species”)

‘Ploceus’ sakalava Sakalava Weaver (“If not in Ploceus, could belong in endemic Nelicourvius, but another genus, Saka, has been proposed… if it proves not to be close to Nelicourvi [Weaver]”)

Hello Andrew

Seems to be a good book. :t:

'Falco' zoniventris can be included into subgenera Cerchneis of genus Falco (in my opinion) and 'Ninox' superciliaris is (I believe) sister taxa of Athene cunicularia so it merge into Athene.

If we use Saka and Nelicurvius for malagasy weavers so we should
fully review the genus Ploceus to splitting into more genera (Xanthoplectes, Othyphantes, Sitagra etc...)

andrew147 said:

‘Falco’ zoniventris Banded Kestrel (“Large Falco genus may deserve splitting into more genera”)

Click to expand...

Cerchneis it would be. (Albeit the most divergent species in the group.) I'm unclear why the genus would need splitting, however.

‘Sarothrura’ watersi Slender-billed ‘Flufftail’ ("Lemurolimnas")

Click to expand...

Genetic data are lacking. Based on morphology, Livezey 1998 (used to be available [here], but not any more; can be read [here], see Fig. 3 on p. 2093) recovered this species deeply nested within Sarothrura.

‘Ninox’ superciliaris White-browed Owl (Authors say no genus name “has ever been proposed for this species” – however, Cephaloglaux Kaup, 1852 seems to apply)

Click to expand...

Was indeed recovered as the sister group of Athene cunicularia by Wink & Sauer-Gurth 2004 [pdf], albeit based on a single cytochrome b sequence of undeclared length, which remains unpublished.
Cephaloglaux Kaup 1852 [OD] was proposed for three species, Strix superciliaris Vieillot 1817 [OD], Athene jacquinoti "Hombron & Jacquinot" = Bonaparte 1850 [OD], and Noctua variegata Quoy & Gaimard 1830 [OD]; type species not fixed in the OD: it could apply, assuming that no-one designated one of the two other species as the type. On an admittedly rapid search, I couldn't find any subsequent designation. This name was quite rarely cited, even as a synonym.

‘Zoonavena’ grandidieri Madagascar Spinetail (“Presumed congeners [give] the genus a rather unlikely distribution; research should test for true relationship or convergence, and resulting possibility of different genera, among the three species”. NB: Z. grandidieri is the type species of Zoonavena so 'Zoonavena' would only apply to the aforementioned 'presumed congeners').

Click to expand...

Data are indeed lacking.

‘Mirafra’ hova Madagascar Lark (“recently shown to belong with…Eremopterix. Given extreme plumage distinctiveness…might merit its own genus”)

Click to expand...

It is embedded in Eremopterix in the result of Alström et al. 2013 [pdf], which means that giving it a separate genus would probably require splitting Eremopterix further. (But, to be frank, this result was unexpected enough that I'd really prefer to see it confirmed by other independent sources...)

‘Ploceus’ nelicourvi Nelicourvi Weaver (“[Ploceus] probably requires subdivision… Nelicourvius was originally established for this species”)

‘Ploceus’ sakalava Sakalava Weaver (“If not in Ploceus, could belong in endemic Nelicourvius, but another genus, Saka, has been proposed… if it proves not to be close to Nelicourvi [Weaver]”)

Click to expand...

(Indeed Nelicurvius Bonaparte 1850 [OD]. Saka is by Roberts 1947, in [this].)

Weavers are a complicated matter.
In genetic data, two main groups of Ploceus are quite clearly visible, the Asian species on one side (P. manyar, P. bengalensis, and P. philippinus, plus P. megarhynchus based on an nd2 sequence; for P. hypoxanthus, there is just a single nd3 sequence in GenBank, not associated to any published analysis, and which is part of a data set that behaves somewhat strangely: this sequence places the species closer to the other Asian species as well, but I'd be much less confident about it), and the African/Madagascan species on the other. These two groups do not seem to be sister. The type of Ploceus is philippinus, so I think it quite likely that this name will at some point have to be restricted to the four Asian species. Additionally, the African group is probably not distinct from Malimbus Vieillot 1805, at least as currently understood: Malimbus nitens is nested within it with high support based on nd3 + atpase8-6 (cf. Warren et al. 2012 [pdf], but beware that the tree on Fig. 2 was rooted arbitrarily on Malimbus nitens; I have joined a tree based on the same loci with additional taxa, including more distant outgroups; the same data also suggest Ploceus sakalava is basal in this African group); short (300bp) cox1 sequences also suggest that Malimbus coronatus may be closer to Ploceus bicolor than to Malimbus nitens. (But data are entirely lacking for M. malimbicus, the type of Malimbus.)

Are there phylogenetics studies of genus Ploceus ?

Ninox jacquinoti is the type species of Cephaloglaux, no ? (i think..., en fait j'en sais rien).

I know no published genetic work specifically about Ploceus. The data in GenBank are mainly from outgroups used in studies on other birds, and from barcoding efforts.
As I said, the type of Cephaloglaux is not fixed in the original work: each of the three species that were included by Kaup could become the type species if designated as such in a later work. Do you know a work where Athene jacquinoti is said to be type?

As always, thanks for the very informative responses.

l_raty said:

Cerchneis it would be. (Albeit the most divergent species in the group.) I'm unclear why the genus would need splitting, however...

Click to expand...

I'm not in a position to critique their methodology but the divergence time estimates of Fuchs et al suggest lineages within traditional Falco that are older than many currently recognised genera in other families.

Primary divergence of Falco into a core kestrel group ('brown Old World and Banded kestrels') and a core falcon group ('American kestrel, grey kestrels and falcons') = 7.5 Mya

Divergence of F. zoniventris from remaining core kestrels = c. 6 Mya

Divergence of F. sparverius from remaining core falcons = c. 6.5 Mya

Divergence of the 'merlin group' (including grey 'kestrels', red-footed falcons and merlins) from remaining core falcons = c. 6 Mya

Further divergence of the previous group has F. vespertinus + F. amurensis diverging from the rest c. 5.5 Mya, and F. ardosiaceus + F. dickinsoni separating from the merlin lineage (F. columbarius/F. aesalon) c. 5 Mya.

I suppose it's a matter of preference but recognising (at least) Cerchneis seems justifiable and the recognition of 4-7 genera in place of traditional Falco would arguably add greater clarity to the interpretation of falcon diversity and inter-relationships.

Ref: Fuchs, J., Johnson, J. & Mindell, D. (2014) 'Rapid diversification of falcons (Aves: Falconidae) due to expansion of open habitats in the Late Miocene' Molecular Phylogenetics and Evolution 82: 166-182.

On balance, Holt & Jønsson 2014 [pdf] found that the cutting date that would produce an age-consistent generic classification in birds without modifying the total number of genera, is 11.402 mya.

For I_raty In my previous post the french sentence : ''En fait, j'en sais rien'' = In fact, I don't know.

For genus Falco, here's how I (personally) reorganized it.

Subgenus Cerchneis
zoniventris, naumanni, rupicoloides, alopex, araeus, punctatus, duboisi, newtoni, rupicolus, tinnunuculus, moluccensis, cenchroides

Subgenus Poecilornis
sparverius

Subgenus Erythropus
vespertinus, amurensis

Subgenus Tinnunculus
columbarius, aesalon

Subgenus Dissodectes
ardosiaceus, dickinsoni

Subgenus Ieracidea
berigora

Subgenus Neofalco
rufigularis, deiroleucus

Subgenus Hypotriorchis
longipennis, severus, eleonorae, concolor, subbuteo, cuvieri

Subgenus Nesierax
novaeseelandiae

Subgenus Rhynchofalco
femoralis

Subgenus Chicquera
chicquera

Subgenus Falco
hypoleucos, mexicanus, fasciinucha, peregrinus, pelegrinoides, biarmicus, jugger, subniger, rusticolus, cherrug

Anyway, everyone is free to use their own classification.

l_raty said:

On balance, Holt & Jønsson 2014 [pdf] found that the cutting date that would produce an age-consistent generic classification in birds without modifying the total number of genera, is 11.402 mya.

Click to expand...

But... if this date was to be consistently applied, it would hugely modify the circumscription of genera, if not the overall total number.

It would mean that a lot of current 'families' were actually equivalent to genera. As examples, the whole diversity within Icteridae, Parulidae, Cardinalidae and Passerellidae is thought to stem from less than 11.402 Mya.

Nomenclatorist said:

For I_raty In my previous post the french sentence : ''En fait, j'en sais rien'' = In fact, I don't know.

Click to expand...

J'avais bien compris
Laurent -

Enfin un francophone, pfiou je désespérais.

andrew147 said:

As examples, the whole diversity within Icteridae, Parulidae, Cardinalidae and Passerellidae is thought to stem from less than 11.402 Mya.

Click to expand...

Yes, these (and more; 14 families in total) all ended up in a single family in Holt & Jønsson's suggested classification.
(But note that they were also in a single family Fringillidae in Sibley & Monroe's classification [which may have been the only real attempt ever to link rank and divergence levels consistently across the classification of birds]. And they are actually in a single family Emberizidae on the 8th ed. of the BOU British List [Harrop et al. 2013 - pdf].)

Enforcing age consistency is not necessarily without consequences...

Where can we find the Holt and Jonsson ' s classification, pleaze ?

l_raty said:

And they are actually in a single family Emberizidae on the 8th ed. of the BOU British List [Harrop et al. 2013 - pdf].)

Click to expand...

Yes, treated as tribes – Icterini, Parulini, Cardinalini, Emberizini (incl New World sparrows) – by BOURC TSC (and Dutch Birding).

Nomenclatorist said:

Where can we find the Holt and Jonsson ' s classification, pleaze ?

Click to expand...

See: www.birdforum.net/showpost.php?p=3054827&postcount=16.

Nomenclatorist said:

Where can we find the Holt and Jonsson ' s classification, pleaze ?

Click to expand...

I went through this carefully when it was first published and there are some fairly 'out there' notions which aren't supported by other recent evidence. For example, just from Galliformes, Arborophila including Tropicoperdix is not monophyletic, Megapodius does not represent a particularly old lineage, Lerwa seems unlikely to be part of an all-encompassing grouse genus... similar examples exist in other groups.

However, some of the suggestions, especially concerning higher level taxonomy, seem intuitive: e.g. multiple genera in place of Turnix, Calidris, Tyto, Ninox, Vanellus, Nyctibius and Falco; multiple families for ancient complex groups (tinamous, plovers, sandpipers); and significant lumping of passerine families (just 42 families).

Personally, I use Vanellus for only one species: Vanellus vanellus.

I am not in favor of using age of taxa to define cut off points for family/genus/species, since no matter what date you use you either get uninformative lumping at some level (does it help anyone to recognize families of 900+ species?), or you get oversplitting of taxa where a bunch of overall similar critters are all put in monotypic genera. The Linnean system predates modern phylogenetics and we should just recognize that, and avoid changing classification unless issues of monophyly/paraphyly/polyphyly come up.

Taxonomic (or classification) changing is a recommandation not a obligation.
Moreover we don't know if Vanellus is really monophyletic? (e.g.)

Nomenclatorist said:

Personally, I use Vanellus for only one species: Vanellus vanellus.

Click to expand...

And what about Vanellus madagascariensis? Goodman (1996) considered it as distinct and extinct.