With the arrival of the Marmora’s Warbler Sylvia sarda in Gwent, perhaps it is timely to review its identification features, and how to separate it from Balearic Warbler Sylvia balearica. Here I adopt the current policies of Birdlife International and HBW in treating sarda and balearica as separate species – though both of these groups base this on Shrihai et al (2001) – HBW write, ‘Forms a superspecies with S. balearica; until recently considered conspecific, but differs markedly in voice (both calls and song, and neither species responds to other’s song), structure, size, migratory behaviour and mitochondrial DNA. Genetic studies suggest that both are more closely related to S. deserticola than to S. undata’. Some debate these changes; it is notable that the Collin’s Guide has not ‘split’ the two, even though it has with other closely related species, like Eastern S. crassirostris and Western S. hortensis Orphean Warblers. ‘Marmora’s Warbler’ exists on the British list as of indeterminate race: the BOURC checklist comments: ‘Marmora's Warbler…vagrant, race undetermined but unlikely to have been other than nominate sarda Temminck’ – I do not know if the BOURC TSC are currently examining the taxonomic status of Marmora’s or Balearic Warbler. There has been discussion of the five British records of ‘Marmora’s Warbler’ (1982, Midhope Moor, South Yorkshire; 1992, Spurn, East Yorkshire; 1993, St. Abb’s Head, Borders; 2001, Scolt Head, Norfolk; 2001, Sizewell, Suffolk) and it has been an opinion held by many that the subspecies involved in these records is sarda. Below, I lay out my own views on the identification criteria and hopefully this might generate discussion.
For many birders visiting the Balearic Islands, one of the key species to see has long been ‘Marmora’s Warbler’ Sylvia sarda, and most make contact with it on Mallorca, where the local form balearica is often seen up the Boquer valley, along the Casas Veyas and Formentor peninsula or on the eastern or southern headlands, e.g. Cabo Blanco. Following the taxonomic changes, many now find they have only seen Balearic on their trips to Mallorca; Marmora’s the nominate form, is to be found on Corsica, Sardinia and smaller islands off western Italy – in my experience on Corsica it is fairly localised, prefering areas of hillside maquis and garrigue. Balearic is largely sedentary, with some post-breeding dispersal to mainland Spain; some Marmora’s move to coastal W Italy, NE Algeria, Tunisia and NW Libya.
When dealing with sarda and balearica (as indeed it is with all Sylvia warblers), the first port of call is of course Shirihai et al, and once you have managed to penetrate the text there is much to digest – the sections to read are ‘2. Field Identification’, ‘2.3 Separating the Two Allospecies’, ‘Voice’ and ‘5. Allospecies Taxonomy’. In places, Shirihai et al is repetitive, but you will gain an idea of a lot of the differences between the two ‘species’ (and from other similar ‘species’). Over the past ten years I have had field experience of the two forms (sarda on Corsica and balearica on Mallorca) and have examined specimens of the two at the NHM at Tring. Shirihai et al focus on five elements useful in separating the two forms: 1. Voice – useful when heard by an experienced birder, described well (or even better recorded); 2. Structure – helpful; 3. Leg colour; 4. Underpart coloration, but not helpful in juvenile (or first-winter?) plumage; 5. Bill colour. I shall deal with each of these in turn, and these link with a number of images on the task site or inserted. Below is a summary of the characters which I feel will separate them.
Marmora's Warbler, France, Barcaggio / Corsica 05/05/06 © Willy Raitiere
Both (sub-) species have characteristic calls and song. In the context of a vagrant, by far the most helpful is the call: on sarda the call is a rich (Shirihai et al call it ‘guttural’) ‘trak’ or ‘tchak’ – reminiscent of European Stonechat Saxicola rubicola; in comparison, I find that balearica has a slightly less strong or defined call, a stuttering and softer, rolling (ie, with a notable ‘r’ in it) ‘trret’ or tsret’, more like a sparrow – note that the recordings commonly available by Roche (and on Birdguides) are of sarda. The songs are also subtly distinct, with sarda having a song that recalls Spectacled Warbler S. conspicillata, a ‘short, twittering warble, slightly higher in pitch and, to my ears, a series of sweeter notes than balearica [sometimes ending with a long repetitive trill’ as described by Shirihai et al]; balearica is slightly more like Dartford Warbler S. undata in tone, with grating notes and is less flowing and sweet, though it too can end with a long rolling trill. Shirihai et al report that in playback experiments neither form responds to the others song, and I can confirm that with Arnoud van den Berg, we played balearica to sarda in Corsica and got no response. Marmora’s can be heard here - http://www.xeno-canto.org/europe/browse.php?query=or:7886.00
It is likely that a vagrant will call, and indeed the only bird I have seen in the UK at Sizewell called a hard ‘tchak’.
2. Size and structure
Though it is very unlikely you will see the two forms together in the field, field views indicate that sarda is a slightly larger bird (with, on average, a slightly shorter and broader-based bill), a sturdy body and a relatively shorter and stronger tail – individual tail feather width of Balearic is narrower than on Marmora’s. My impression of Balearic Warbler, on the other hand, is of a smaller bird, with a spiky bill, a rounder weaker body and longer, narrower tail – more reminiscent of a Dartford Warbler (or even weedier like a Long-tailed Tit). My field observations have by-and-large been confirmed when examining skins; Shirihai et al give the following measurements (from various sources), confirming that Marmora’s is a slightly larger bird:-
Wing (total range) sarda 50-60mm (mean 53.5-56mm) balearica 46-54mm (mean 48.8-50.8mm)
Tail sarda 53-66 (mean 53.5-56mm) balearica 51-60mm (mean 53.8-55.8mm)
Bill length (Shirihai) sarda 12.5-15.9mm (mean 13.7mm) balearica 12.3-15mm (mean 13.6mm)
Bill depth sarda 2.7-3.5mm (mean 3.0mm) balearica 2.6-3.1mm (mean 2.8mm)
3. Leg Colour
The Collin’s Guide illustrates and points out differences in leg colour between Marmora’s and Balearic, ‘orange-brown’ and ‘brownish orange’ respectively; however, I am less happy about the leg colour being a discriminating. Shirihai et al state that this is largely only helpful in dealing with young birds, with sarda having ‘mostly dark greyish brown legs (rarely dull orange-brown)’ contrasting with the ‘typical pale orange to orange-brown of balearica’; again, personally, in the field I feel that the leg colour of young birds is pretty similar.
See: Marmora’s - http://www.netfugl.dk/pictures.php?id=showpicture&picture_id=31494; Balearic - http://www.netfugl.dk/pictures.php?id=showpicture&picture_id=17753
Rather than describe the plumage of each form in detail, in the context of their ID it is probably better just to examine differences between the two.
The upperpart colour and tone of the two forms is generally the same (indigo) slate-grey. Depending upon age and sex there are more or less brown tones, with females of both having a brown element to the mantle and wing edges – though I feel that fm sarda has browner upperparts and more ochreous brown edges to the remiges than balearica. However, it is largely the underparts that show the greatest difference, with sarda having darker and greyer underparts, opposed to balearica, which is paler below and has more pinkish or brown hues - in this respect, sarda has a darker overall tone than balearica, with the upperpart colour coming round onto the chin and throat (sometimes marked with small white feather tips, as in Dartford) and upper breast. On male sarda in spring the strength and extent of this throat colour may leave a small white area on the malar closest to the bill – seen from the front this looks like a Mexican moustache - http://www.netfugl.dk/pictures.php?id=showpicture&picture_id=6622; male balearica have a paler grey breast, which washes only lightly onto the lighter throat, and therefore lacks the small white spots on the malar - http://www.netfugl.dk/pictures.php?id=showpicture&picture_id=17754 (I am doubtful as to whether males of either form really ever has the contrastingly white throats shown by the illustrations by Alan Harris in Shirihai et al). The grey colour of the breast on sarda extends down the flanks, where it can sometimes have a brown tinge (most likely on younger males); on balearica the tone is paler and the flanks often have a distinct pinkish brown hue. I also feel that the rump and upper tail is darker on sarda than balearica, which has a paler pastel-grey fringe to the central tail feathers. The dark loral area may be also more extensive onto the side of the crown in male Marmora’s
Male, Corsica (BS) – note the moustache
Females of both are pretty similar above, though as I have said already, I often think that sarda is slightly browner above with more ochreous fringes to the remiges (notably the tertials). The underparts are subtly different, with the lower throat and breast of sarda having darker, greyer hues than the pale pinkish brown-grey of balearica – this colour is most obvious along the flanks.
Marmora’s: http://www.netfugl.dk/pictures.php?id=showpicture&picture_id=6606, http://www.netfugl.dk/pictures.php?id=showpicture&picture_id=24012
Balearic Warbler: http://www.netfugl.dk/pictures.php?id=showpicture&picture_id=33908
5. Bill colour
Shirihai et al hint that there is less black on the tip of the lower mandible on balearica, but I do not think this is borne out by my field experience or photographs. Sometimes, perhaps, it is possible to see a reddish flush at the base of the bill on breeding male balearica which I have never seen on sarda (the crimson red bill painted by Lars Jonsson in ‘Birds of Europe’ may be a touch too much).
Shirihai, H., Gargallo, G., Helbig, A., and Cottridge, D, 2001: Sylvia Warblers – identification, taxonomy and phylogeny of the genus Sylvia; Christopher Helm, London