Seeking Hebridean Song Thrush

Sometimes a story is too big to begin at the beginning. When, after breakfast, I walked out of my accommodation in Castle Bay on the Outer Hebridean island of Barra in mid April 2015, I knew that today was going to be the one that concluded hundreds of hours of wondering. A little over a century before — and on close to the same date — William Eagle Clarke shot a pair of Song Thrushes in precisely the same spot. The eponymous settlement snakes around the edge of a high bluff and overlooks Kisimul Castle perched on a rock in the middle of the bay. The fortress dates from the fifteenth century and has aged well. A bright blue and white saltire fluttering from a defiant flagpole suggested that proprietorial rights were still being maintained. I had come all this way to see the living relatives of two dead thrushes and I felt the poised hand of history.

William Eagle Clarke was the first to declare in print [1] [] that Song Thrushes inhabiting Scotland’s Outer Hebrides were sufficiently different from other Scottish (and British) Song Thrushes to warrant separate rank. Although not worthy as a full-blown species, they showed a measure of uniqueness that set them apart and qualified them as a subspecies, Turdus philomelos hebridensis. Eagle Clarke defined their characteristics but also acknowledged that other naturalists had long known the bird to be different, such as Robert Gray, who in Birds of the West of Scotland (1871), stated (p.76) that when on North Uist in 1868 he ‘remarked particularly the unusually dark colour of their plumage — the birds being very unlike those brought up in cultivated districts.’

Nowadays there is a tendency to belittle the observational skills of our forebears []. Modern optics and digital photography are certainly revelatory but we tend to forget that earlier ornithologists had, if anything, better equipment because they shot the birds that they considered unusual. Hence visual acuity — the ability to tell that something was out of the ordinary — was followed up with hard evidence. We should read such history with a view to revisiting the experience and set ourselves, not above early pioneers, but alongside them. They attained objectivity on their own; their footsteps are our guide.

Eagle Clarke was the keeper of the Natural History Department of the Royal Scottish Museum in Edinburgh (now part of the National Museum of Scotland) and blazed a trail that launched ‘island going’ as a means of discovering scarce migrants and understanding the influence of weather on bird migration. In other words, he did the heavy lifting in his era. What we inherit from him is a laboratory description of Hebridean Song Thrush that has a close relationship with the truth. He wrote:

‘The Hebridean Song Thrush … is decidedly darker in its plumage generally than any of the British or Continental representatives of the species. The mantle and wings are dark (clove) brown, the head slightly redder, and the rump and upper tail-coverts dark olive. Thus the upper plumage of the Hebridean bird more resembles that of the continental race than that of its British cousin. The most striking feature of the under surface, and indeed of the plumage generally, is the multitude of intense black, ovate spots on the throat, chest and abdomen; while the buff, which is confined to the throat and chest, is very pale as compared with Song Thrushes from other areas. The flanks are pronouncedly streaked with greyish-brown, and show little of the buff which is much in evidence on the flanks and breast of other Song Thrushes. The buff of the under wing-coverts is richer (redder) than in other specimens with which I have compared them. … Though the native [Hebridean Song] Thrushes are permanently resident in, and abundantly and widely distributed over, the glens, moorlands, mountains and sea-cliffs of the Outer Hebridean Islands, it is possible that in certain districts all do not belong to the aboriginal race. Of late years, extensive woods have sprung up in the grounds of Stornoway Castle [on Lewis, the most northern island in the Outer Hebrides] and some of the Song Thrushes resorting to them for nesting may be colonists from the Scottish mainland. Indeed, I have examined one such specimen obtained by Mr Kinnear in the Stornoway woods on 11th July 1906 … which certainly belongs to the British race [2]. Mr Kinnear also found the Mistle Thrush established in the same woods as a nesting species — another recent colonist to the islands. It must also be remembered that Scottish and Continental [Song Thrushes] do not quit them until the nesting season of the native birds has well set in.’

Eagle Clarke’s pronouncement was made in 1913. Apart from its intrinsic value in documenting the look of T.p.hebridensis, the admittance that some Song Thrushes breeding on the Outer Hebrides could be outlanders was tantamount to hedging a bet. Reading between the lines, for anyone wishing to see the real McCoy, any visit to the home range should be directed at the epicentre. For me, that meant Castle Bay in the breeding season and concentrating only on breeding pairs.

Eagle Clarke died in 1938 but his investment and faith in T.p.hebridensis endured and was endorsed in 1940 in the publication of The Handbook of British Birds [3]. The Handbook’s chief author, Harry Witherby, stuck to Eagle Clarke’s words. However, time having happened, further nuggets of knowledge were added to the cairn. Under ‘Field-characters and General Habits’ the Handbook stated: ‘Upper-parts and flanks decidedly darker than typical race [T. p. clarkei: named in Eagle Clarke’s honour and is the form occurring from Ireland east to the near Continent] and under-parts more thickly spotted render it fairly distinct, but as various reliable observers record flocks of noticeably dark (but apparently small) thrushes, of problematical origin and most unlikely to be all hebridensis, as passage-migrants or winter-visitors to mainland of Great Britain, it cannot be safely identified in the field outside its normal habitat.’ A possible slender reed for a bird in the hand was a statistic given under ‘Measurements and Structure’. This read: ‘As in British Song Thrush, but bill rather larger on average, viz. 22–24mm. (12 measured).’

Farther on, in a footnote beneath ‘Description’ the water was muddied further: ‘Song Thrushes from Skye [part of the Inner Hebrides] seem not separable [from T.p.hebridensis] but birds from other Inner Hebrides and parts of W. Scotland and Kerry are intermediate in colour between the British [T.p.clarkei] and Hebridean forms.’ With the mention of County Kerry a local story had, in my mind, widened to include global proportions. Then, on 1st January 1963 a Song Thrush caught and measured at Roscrea, County Tipperary was considered to show all the characters of T.p.hebridensis. Its entry in the Irish Bird Report of that year (pp.30) states: ‘plumage compared directly with [T.p.clarkei] from which it differed as follows: mantle and back much darker brown, rump and upper tail-coverts rather greyish and quite distinct from the rest of the upper-parts; under-parts much more spotted, the spots larger and darker; flanks dark smoky-brown. Bill 23.5mm. There was a striking contrast to two [T.p.clarkei] caught at the same time.’

Plate 1. Hebridean Song Thrush (Turdus philomelos hebridensis). Bird A. This and other Hebridean Song Thrushes (see below) were photographed within a 3km radius of Castle Bay, Barra, in mid April 2015. The instant and overriding impression was of a cool-coloured version of a ‘conventional’ Song Thrush (whose brown upperparts look ‘pancake brown’ — Martin Gray). On Hebridean Song Thrush the sides are murky and the big dark spots on the chest bleed into foggy flanks. The photographs were taken in overcast light. Nonetheless — as in life — colour cast varies in line with the background. When a bird moves and poses against a different backdrop, tints change. In an attempt to achieve a faithful portrayal, a neutral grey background has been used. If anything, digital reproduction exacerbates the situation because of variation between monitors; what looks pure brown on one screen may appear closer to grey-brown on another. All image processing was carried out on a colour-corrected monitor. Several images have been reversed for easy comparison.

Watching birds, once you discover the fascination, is a lifelong process. You do not complete your career as though it is some kind of driving test. Birds are the career. The possibility that a diagnosable-in-the-field subspecies of Song Thrush from the Outer Hebrides might occur as a migrant or winter visitor to Ireland had not registered on my radar until I bumped into a Song Thrush with a chocolate-coloured back and spotted undersides redolent of a Dalmatian dog. I was transfixed. I saw the eye-opener in autumn 1998 on the coast of Donegal in the northwest corner of Ireland. Location and timing suggested a migrant. Moreover, if the bird arrived from close to due north, the nearest landmass from which it might have departed was Barra, Outer Hebrides. Shameless speculation, of course. When, over subsequent autumns, I encountered two other potential T.p.hebridensis in west Donegal, the bug had well and truly bitten.

Curiosity is a wellspring that is never extinguished until irrefutable evidence proves a point. It turned out that not only did I want to get to the bottom of the status of T.p.hebridensis in Ireland, so too did Scottish ornithologists. In particular, Bob McGowan, filling Eagle Clarke’s shoes as curator of the bird collection at the National Museums of Scotland, got in touch. You can imagine why.

At this stage there is no need to keep hammering away at a nail whose head is already flush with the wood. Although leads and events did not move quickly, they steadily achieved critical mass. In December 2003 I saw the touchstone of the debate in a museum tray in Edinburgh. Eagle Clarke’s type specimen struck me as it did him: dark-backed and heavily marked below.

Type specimen of Hebridean Song Thrush (Turdus philomelos hebridensis). Image reproduced from Scottish Naturalist, 1913, Plate 1 (facing page 53).

But did it ring a bell with what I had seen in Ireland? Perhaps. I wanted to be prepared to face a living Hebridean Song Thrush in my own time and know what I beheld. While Eagle Clarke’s words mattered, sincerity demanded definitive field marks; anything less would be myth-mongering.

Plate 2. Hebridean Song Thrush (T.p.hebridensis). Bird B. A capped effect was obvious when the head tilted; a rufous wash was discernible across the folded flight feathers and greyness on the rump contrasted with ‘clove-brown upperparts’ (sensu W.Eagle Clarke). All of these features can, with varying intensity, be shown by Song Thrush (T.p.clarkei).

Trying to maintain a benevolent sanity was not easy. Information on some Hebridean web-sites stated that Song Thrushes ringed on the Outer Hebrides had been recovered in Ireland. I contacted the British Trust for Ornithology and they put the record straight. Of eight Song Thrushes ringed in the breeding season on the Outer Hebrides (more than likely, T.p.hebridensis) seven were later recovered still on the Outer Hebrides. The eighth bird had come from Nairn in Aberdeenshire where it was ringed as an adult on 16th May 1984 and was recovered on Lewis on 3rd January 2000. In all probability it was T.p.clarkei, a typical winter visitor. Finally an identification article dealing with T.p.hebridensis and other endemic Scottish subspecies appeared in Scottish Birds [4] and was republished in summary form in Birding World [5]:

Plate 3. Hebridean Song Thrush (T.p.hebridensis). Top row: Bird C; middle row: Bird D; bottom row: Bird E. Any species whose plumage consists of colour washes of different strength as well as a multitude of spots, is bound to vary. Bird C has a warm glow across the chest; Bird D has bright underparts with little sign of dusky flanks; Bird E is, however, pretty much ‘on the money’ and shows a good suite of T.p.hebridensis characters.

‘Although first described in 1913, we are still largely ignorant of the status, distribution and movements of Hebridean Song Thrush. Probably confined to the Outer Hebrides and the Isle of Skye, with a population estimated at not less than 500 pairs and not more than 4,000 pairs (B. Rabbitts, A. Stevenson, A. Currie & R. McMillen pers. comm.), T. p. hebridensis is generally believed to be largely sedentary. A proportion move away from their breeding grounds between July and February and it has been suggested that some may move to Ireland or western mainland Scotland, though proof is lacking. Some Continental Song Thrushes T.p.philomelos visit these islands in winter, as do some mainland Scottish British Song Thrushes, T.p.clarkei. Birds from the Inner Hebrides and parts of western Scotland are described as intermediate between T.p.hebridensis and T.p.clarkei and several specimens from Islay at the National Museum of Scotland, Edinburgh, appear to show mixed characters (see McGowan et al. 2003).

Plate 4. Hebridean Song Thrush (T.p.hebridensis). Top left: Bird F; top right: Bird G; centre left: Bird H; centre right and bottom: Bird J. Having looked at Song Thrushes (T.p.clarkei) in the flesh and as museum specimens — chiefly in the extensive collection held at the Ulster Museum — none of these four rang any alarm bells that said ‘Hebridean’. Yet because they were breeding within the core range of T.p.hebridensis they were, de facto, Hebridean Song Thrushes. Geography aside, I felt that they were otherwise similar to some T.p.clarkei.

Six features are useful in differentiating T.p.hebridensis from T.p.clarkei and T.p.philomelos, [the subspecies of Song Thrush occupying Europe east of T.p.clarkei] and identification should be based on a suite of these characters, not just one. Ventrally, these are: 1. bold, almost black spots; 2. flanks smoky, greyish-brown rather than buffish; 3. underwing-coverts rich, rufous buff and darker than other races. On the dorsal surface: 4. forehead, crown and mantle dark brown with less rufous tone than clarkei; 5. rump and uppertail-coverts with grey wash; and 6. dark markings of face and throat generally bolder. Regarding the last feature, it is worth emphasising that the forehead and crown are noticeably darker in hebridensis when directly compared with the other two races. Most Hebridean Song Thrushes should be separable in the field given good views. Bold underparts spotting alone should not be relied on: the grey wash on the rump and uppertail-coverts, smoky-brown flanks and richer rufous-buff underwings are important identification features.’

Having been there and done that, I was sceptical of some of the criteria advanced for identification. Rump colour on all Song Thrushes varies between grey and brown. Furthermore, individuals with brown rumps can, with a change of angle, become grey-rumped. Another novelty was the allusion to a ‘capped’ effect for T.p.hebridensis born out of ‘the forehead and crown [being] noticeably darker’. All Song Thrushes that I have looked at in Ireland and Britain are, to a degree, capped. The strength of the feature is linked to head angle. Especially when a Song Thrush cocks its head to one side, the crown can resemble a chestnut bonnet. However, to be fair to the published article, the authors did say that the darker cap of T.p.hebridensis was best appreciated only when directly compared with other races. Thereby making the criterion useable only when comparing specimens?

Plate 5. Song Thrushes (T.p.clarkei). Four different individuals from Ireland. The two left-hand birds were in coastal County Down in January. They may have been local or winter immigrants from across the Irish Sea. The two right-hand birds were on Inishbofin, County Galway, and were photographed in March. Both were almost certainly local breeding birds (the far right bird was singing and defending territory). Although plumage appearance linked to age is not a factor in the T.p.hebridensis versus T.p.clarkei debate, it is useful to point up some field marks that denote age. I am grateful to Neville McKee for drawing my attention to the following: ‘the far right bird has pointed tail feathers indicating that it is a first-winter (hatched during the previous summer); the far left bird is also a first-winter because the large buff-coloured tips to its greater coverts connect with the feather shaft. The central two birds are, almost certainly, adults.’ Note that certain features manifest on three of the four are cited as applicable in lending support to identifying T.p.hebridensis. Such as: the strength of the rufous cap (far left bird); dusky flanks (left of centre); grey rump (right of centre). The right-hand individual is included as a ‘control’ because it shows a more typical brown rump, albeit with a hint of grey.

I tried harder with Song Thrushes in Ireland. Inishbofin, west of County Galway, has a breeding population and receives many immigrants in autumn. During winter, the species is almost the commonest songbird with a hundred or more on the island. Although it felt increasingly like stepping carefully through a minefield, I deemed two individuals worthy of scrutiny. They were dark, earthy brown above with heavily lined flanks and one showed next-to-no-sign of the yellowish ‘glow’ that, on a typical Song Thrush, permeates the plumage from chest to flanks. I emailed images to Bob McGowan. Understandably, he was reluctant to say that I had hit pay dirt. Unless I had a bird in the hand and morphed into a man in a white coat prepared to take DNA samples, there was always a need for a better, more watertight candidate. The quest was less like the hunt for Higgs boson and more like Goldilocks tasting for porridge that was just right. I packed my bags for Barra.

I arrived in Castle Bay at dusk. Too late to see Song Thrushes, but not too late to shortlist the best habitat. In driving rain the airport bus passed a collection of houses surrounded by thickets of willow and alder through which a boardwalk spanned marshy ground and ended at a soccer pitch. The spot was called Horvie and it looked like Hebridean Song Thrush heaven. I made a bee-line there straight after breakfast. After an overnight deluge, foliage bristled with grounded Willow Warblers, Chiffchaffs and Blackcaps. For once I ignored their leaps and scanned low down. I could hear Song Thrush song in the distance.

I knew that my store of research would be ignited by a single spark. The initial effect would not be assembled from a distillation of all that went before; it would be a sudden hit, like a baby’s first cry. I clocked the silhouette as it halted among horsetails masked by a filigree of branches. Somehow I had not reckoned on communing with a silhouette. With a few hops the shadow flew up and landed on the wooden rail of the boardwalk. Colour at last. Colour was the thing. In fact it was the only thing that struck me. But what’s so ‘only’? I was dumbstruck. The effect was akin to looking at a Song Thrush through smoked glass (Plate 1, Bird A). Before rebooting my brain in the hope that it would absorb detail, a voice in my head screamed, ‘you have never seen anything like this before.’ The Holy Grail was a dusky dream. Grey trees and dripping raindrops cast a neutral light and rendered a Mistle Thrush palette on a Song Thrush. Depending on angle and background, the spectrum swung between grey and brown. Plumage small print was best described in negative terms: not particularly dark brown above; breast spotting not overwhelmingly profuse; and the rump, whilst tinted grey, did not contrast markedly with the back. Shocks aside, this had to be a Hebridean Song Thrush.

The bird’s uniqueness was more than I bargained for. It was as though bright tawny tints on the chest had been turned down and replaced with a grey fog that hung over the flanks. In certain lights I even felt that the plumage looked oiled (it wasn’t). Nonetheless, the lower abdomen and under-tail coverts were lightly marked, more so than on several Irish Song Thrushes with which I was acquainted. I could almost recite Eagle Clarke’s 1913 description. If what I was seeing was what he saw, the essence of his words still made sense: ‘…the buff, which is confined to the throat and chest, is very pale as compared with Song Thrushes from other areas. The flanks are pronouncedly streaked with greyish-brown, and show little of the buff which is much in evidence on the flanks and breast of other Song Thrushes.’ During the next hour the bird’s mate appeared and then an interloper. All conformed to the same prototype. I decided that, rather than dwelling on plumage nitty-gritty, a holistic approach might be the key to pigeon-holing T.p.hebridensis.

I was elated. If I had booked to fly home later that day I would have railed like a reporter who had scooped an exclusive. My headline might have read, ‘Hebridean Song Thrush is monochrome plaid not chequered tartan.’ But I would have been wrong. Over the course of the next five days I sought out Song Thrushes wherever I went. The sun shone, the wind dropped and the birds came out to play. The more I saw the more I despaired. The litmus test was the question, ‘if I saw that in Ireland would I be able to tell it from a Song Thrush?’ Time after time the answer was no. Even when I answered ‘yes’ I heard the lid of Pandora’s box creaking open and reminding me about those deep-hued singletons from Donegal and Inishbofin. Were they simply dark locals? Except that, based on the first flush of the genuine article, ‘dark’ was no longer the operative word. ‘Dusky’ and ‘swarthy’ were more apt. Instead of clarity I sensed confusion. I did not want to misconstrue what I observed and thereby dramatize a technical story by fudging it. I reverted to the role of a reporter, albeit with an amended brief — get the documentary evidence first and run the analysis later. That way, any assertions would be supported by photographs.

Thus far, I have skipped two criteria of potential significance: the colour of the underwing-coverts and vocalisations. Well aware that T.p.hebridensis exhibits ‘richer (redder)’ [1] underwing-coverts than other Song Thrushes, I made an effort to check. Although I caught flashes when birds flew, I did not succeed in making a definitive judgement. All I gained were impressions and one unsatisfactory image [Plate 4, at bottom]. My gut reaction was that some individuals did possess quite rufous underwing coverts; others less so. However, I also noted rufous underwing-coverts on one Song Thrush seen on Inishbofin, County Galway. I heard lots of song. I took this as a keystone indicating that the singer was a local breeder, ergo T.p.hebridensis, and not a migrant tuning up before heading east. Because I admire Song Thrush in full voice I like to take in the performance. Compared to many songsters I have listened to in Ireland, no obvious difference struck me. If I was to split hairs and compare Inishbofin to Barra, I felt that vocabulary was less varied and the volume was slighter for those I heard on Barra. But such comparisons are meaningless because many factors influence song delivery, never mind the differing singing talent of individuals. One Barra singer (Plate 3, Bird D) can be heard here: [].

‘Subspecies, what subspecies?’ I don’t want the take-home message from my trip to Barra to undermine the inviolability of T.p.hebridensis as a distinct entity. Most outcomes to puzzling phenomena are arrived at by following a single line of argument through selected perceptions to a solution that stands up to scrutiny. At least at this stage I no longer feel like the man who knows least about the subject. If Hebridean Song Thrush exists within a nutshell, then there is a need to pick out its meretricious details. In 2007, the Scottish Ornithologists’ Club published The Birds of Scotland [6], acclaimed as ‘the most detailed account of Scotland’s birds ever published.’ The encyclopaedic work contains the most recent summary and analysis of Scotland’s nine endemic subspecies, including T.p.hebridensis (curiously, no summary of field characters is given for any endemic subspecies).

The Birds of Scotland entry contains the following passage (pp. 1151), ‘T.p.hebridensis represents the end of a cline of increasing colour saturation, with birds of this population darker and more richly coloured in comparison to the paler clarkei elsewhere in Britain and the near continent. This trend accords with Gloger’s Rule under which populations in wetter environments that tend to comprise birds that are darker than those in drier environments, which are paler and more faded in appearance. Thus clarkei has less saturated colours than hebridensis, and nominate philomelos from drier continental Europe is even paler.’

To make the entry in The Birds of Scotland more approachable, a few words of explanation might help. A stable and widespread population of a species whose range is continuous may, through years of natural selection, become ‘graduated’ due to the influence of local climate. Such graduation is called a cline and is especially true for climatic factors such as temperature and humidity [7]. Members of species that live in colder climates frequently have larger bodies than their counterparts living in warmer climates. In relative terms, bigger bodies have a greater surface area and are better able to conserve heat. This principle is known as Bergmann’s Rule. In a remarkable demonstration of how quickly evolution (and Bergmann’s Rule) can occur, introduced populations of House Sparrows resident in cold latitudes of North America commonly have larger bodies than their relatives living in warmer southern climates [8]. A similar example of Bergmann’s Rule is found among redpoll populations. However, the speedy rate of adaptation shown by House Sparrows could have implications if any stereotyped population finds itself, through expanding its range, in a climatic zone that is substantially different.

Gloger’s Rule expresses another ‘gradient’ found within the environment. This time the gradient is mildness and humidity. Under those influences, feathers darken. The benefit of darker plumage lies in increased resistance to degrading bacteria. Plumage in humid environments has a greater bacterial load and humid environments are more suitable for microbial growth. By incorporating, within the feather, deposits of a ‘tougher’ type of melanin (melanin comes in three types of which dark brown to black eumelanin is more resilient to wear) plumage becomes more difficult to break down [8]. Although Gloger, a German zoologist, noted the link between climate and avian plumage colour in 1833 [9], Pallas remarked on the same co-variation over twenty years earlier, based on his experience across the Russian Arctic [11].

Let’s get back to Hebridean Song Thrush. A good way to think of a cline based on colour saturation is to imagine a set of paint cards. Pooling all the literature, T.p.hebridensis is the swarthiest of all Song Thrushes. By accepting Gloger’s Rule it is reasonable and logical to believe in the permanence of an environmental impact on the plumage characteristics of T.p.hebridensis vis-à-vis T.p.clarkei. But how much variation exists? Several questions arise. Has there always been variation, right back to the time of Eagle Clarke and his forebears? Were those ornithologists selective in latching on to the ‘best of breed’ and ignoring others than missed the cut? Did our forebears indulge in cherry picking and finish up allowing the tail to wag the dog?

Plate 6. Song Thrushes (T.p.clarkei) and, in yellow box, Hebridean Song Thrush (T.p.hebridensis). Just days before travelling to Castle Bay on Barra, Outer Hebrides, I scrutinised breeding Song Thrushes on Inishbofin, County Galway. All three birds in this montage were photographed within a week of each other. The Hebridean Song Thrush is repeated from Plate 3, Bird D. I feel that because the photographs were taken around the same time and that the light and poses are similar, comparisons are valid. Allowing for ‘standard variation’, the outcome seems to me to amount to a draw.

Although my perusal of T.p.hebridensis specimens in the collection at the National Museum of Scotland in Edinburgh was superficial, I was slightly perturbed by the less-than-classic look of several. It crossed my mind that their recognition could only be achieved on a ‘when in Rome’ basis: if they originated from among the breeding population on the Outer Hebrides then they were Hebridean Song Thrushes. Bird identification by ideology?

Is it fair to say that, outside the Outer Hebrides, incontestable sightings of T.p.hebridensis amount to Scotch mist? Based on The Birds of Scotland there appear to be no watertight occurrences outwith the subspecies home range. In fact, some existing claims are rejected by the book’s authors. Where does this leave references to populations deemed ‘intermediate’ that hail from the Inner Hebrides and other parts of the Scottish west coast [6], Islay [4] and — of all places — County Kerry [3]? Just how much variation is there and how does it affect the overall validity of the subspecies?

Plate 7. Hebridean Song Thrush (T.p.hebridensis) in yellow box; Song Thrushes (T.p.clarkei) from Inishbofin. This particular Hebridean Song Thrush has not featured before in the article so, to differentiate it, I will call it Bird K. The Inishbofin individuals seem to fulfil more identification criteria favouring T.p.hebridensis than the real McCoy T.p.hebridensis. The Hebridean Song Thrush was photographed in mid April, as was the Song Thrush to its lower left. The bird at top left was photographed in March. The two bottom images were taken in September and depict opposite sides of the same bird.

A primary impulse is to assume that the bird’s ‘genetic’ status quo has remained unchanged since the days of Eagle Clarke. For sure, some individuals are swarthy and eye-catching — a difference that could be explained in conventional genetic terms if a recessive trait persisted in the Outer Hebrides population. But remember this: ‘of late years, extensive woods have sprung up in the grounds of Stornoway Castle and some of the Song Thrushes resorting to them for nesting may be colonists from the Scottish mainland. Indeed, I have examined one such specimen obtained by Mr Kinnear in the Stornoway woods on 11th July 1906 … which certainly belongs to the British race.’ Unless a DNA SWAT team is dispatched to scour Scotland’s western seaboard we will never know if Eagle Clarke’s words were prophetic and that T.p.clarkei colonists infiltrated the ranks of T.p.hebridensis. If so, the plumage of generations of offspring could reflect a new genetic interlocking and explain the variability that I encountered. If that scenario occurred, the long term implications might ultimately be a return to more widespread swarthiness, given that, theoretically, Gloger’s Rule ought to apply once more? Moreover, if the example of House Sparrows in North America, whose change in size conformed to Bergmann’s Rule if anything is to go by, then Song Thrushes that choose to live in a ‘Hebridean’ climate may manifest change sooner rather than later. Before speculating that the planting, around 1850, of Stornoway woods might have favoured mainland colonists, it should be remembered that the Outer Hebrides (in common with much of Scotland and Ireland) were formerly temperate rainforest, denuded during the Neolithic and Bronze Ages. Given that historical context, there would have been more than enough habitat around 8,000 years ago in the environment among which Hebridean Song Thrushes could have evolved their darker cloak.

Gloger’s Rule and Bergmann’s Rule are just two ‘geographical’ yardsticks that seek to explain patterns of variability in the natural world. A hypothesis that seeks to reconcile colour saturation in plumage with local environmental factors was outlined to me by Neville McKee. Commenting on the swarthiness of a Song Thrush (Plate 7, bottom) that I photographed on Inishbofin, County Galway in the hope that I might have bagged T.p.hebridensis, Neville commented:

‘I smell sea air and salt spray here. I am voting for added mineral salts enhancing colour richness. Other species [such as Meadow Pipits and Coal Tits] that are richly coloured near the Atlantic coast fade the farther inland you look. All the species in question are moulting in areas where the mineral content of their food is of a different spectrum of types than that in central continental zones. The wide mixture of mineral salts found in soil is of a quite different balance between coastal and continental areas. Extreme examples would be sodium and selenium — the latter is scarce in high coastal rainfall areas, hence giving nutritional problems for animals that have a requirement for it, albeit in tiny quantities … as I understand it, the mineral balance of soils progressively changes gradually over many miles, often hundreds, away from the sea. An extreme example is sodium which is so scarce in central continental zones that many creatures, including humans, crave for it and will lick it given the opportunity. Apparently, salted Icelandic fish are a supreme luxury in Central Africa because of the sodium … as regards your encounters with actual T.p.hebridensis on Barra … the evolutionary aspects of Hebridean stock interest me too. While I’m suggesting that nutritional factors may influence plumage, it is perfectly possible that they will promote genetic change. Just because the most westerly birds are at the end of a cline, such a state of affairs does not demand uniformity at the extremes. The intergrades change their proportions along the cline. Theoretically there should be 100% conformity to type at the ends of a cline (as presumably exists across the eastern limit of Song Thrush distribution in Siberia) but there is no requirement for a 100% dark type at the extreme western end because, biologically, the population is small. Hence, maybe 80% or 90% ‘Hebridean-type’ (or whatever it happens to be) would be acceptable, especially as the development of the dark type has only been evolving since the last Ice Age (less than 10,000 years ago). The number of gene mutations required to produce a Hebridean type would not be that many and, given the time span, could certainly be achieved.’

Although getting well out my comfort zone of understanding, I get Neville’s point. His contention is that local effects on plumage appearance are derived through a combination of climate and diet. Nick Watmough, who happens to be both a birdwatcher and a biochemist, delved deeper into the same theme and explained how — theoretically but in line with new research — the process might work in practice. Nick wrote:

‘Epigenetics is a developing area of research. [] The concept is difficult for anyone, especially those of us brought up on Mendelian genetics and Darwinism, to comprehend. In essence, all organisms inherit a blueprint that defines a broad phenotype. But the blueprint can be modified by chemical tags placed on the DNA. This has the effect of switching genes on and off. Importantly, this tagging can respond to environment. So a purely hypothetical example might be that Coal Tits have a gene that influences yellow pigment that is ordinarily switched off but, in response to an increasingly humid environment or salt balance in the diet, the pattern of tagging changes. In this way, a gene may be expressed at a ‘low key’ level. However, to a nineteenth century field naturalist, the difference would be observable and through force of argument, hailed as a distinct phenotype deserving recognition as a sub-species.’

Having seen Hebridean Song Thrush in the flesh, I am back at the beginning. I have ditched the composite portrait that I had previously formulated in my mind. Unless I stumble across a smoky sensation in Ireland, I will have to suspend belief. Gone is the habit of peering at images of Song Thrushes photographed in the breeding season on the Outer Hebrides and conceding that, even though they looked like some Song Thrushes from Ireland (and elsewhere), they were somehow different. Such a conditioned response put me in the company of Pavlov’s dogs.

Plate 8. Song Thrush, Inishbofin, County Galway, October 2008. This adult has (almost) completed its post-breeding moult. The plumage is fresh and new, except for some quills on the face that are not yet fully open. Found in the course of a wet day —often the kind of weather that drops migrants on to Inishbofin — I assumed it was a new arrival. Not having been to Barra and encountered ‘classic’ T.p.hebridensis such as Bird A (Plate 1), I never imagined that a ‘grey and smoky’ Song Thrush might equate with T.p.hebridensis. Given the benefit of hindsight, is this a Hebridean Song Thrush that reached Ireland? I do not think I will ever know. Shuffling its undeniably pro-Hebridean field characters and hoping for an unequivocal result is probably as fruitless as rearranging the deck chairs on the Titanic.

It is difficult to know if authentic examples of T.p.hebridensis represent an inselberg towering above a bed-rock of T.p.clarkei or the debatably discernable end point of a cline stretching along the Celtic fringe of Scotland and Ireland. I am also conscious that a visit lasting five days to a few random locations on one island in the Outer Hebrides is far short of a systematic overview. If, as could be the case, individuals matching Eagle Clarke’s type specimen are in a minority, then there may be a dichotomy in the overall Hebridean population that I failed to appreciate. Which leaves me wondering if my gut reaction of ‘50 Shades of Brown’ is misrepresenting a more subtle truth?

Lewis Namier, a noted historian of Russian extraction, comes close to reflecting my sentiments on the depth of current knowledge pertaining to Hebridean Song Thrush: ‘Historical research to this day remains unorganised, and the historian is expected to make his own instruments or do without them; and so with wooden ploughs we continue to draw lonely furrows, most successfully when we strike sand.’


Philip S. Redman spent six months on Tory Island, Co. Donegal, in 1954. He was there continuously from the middle of August until the middle of November. During November, he encountered several dark Song Thrushes:

“One further passerine species requires detailed comment and that is the appearance of large, dark Song Thrushes in November. A bird considered to be of the form T.ericetorum hebridensis [T.p.hebridensis] was observed at close range and in good light on the 14th. It showed very dark brown upperparts and very dark and heavily spotted underparts. It agreed very well with a skin of this race in the expedition’s collection [Philip Redman had been loaned a specimen of T.p.hebridensis by Dr. Harrison who had a specimen in the Harrison Museum at Sevenoaks in Kent]. On the wing it appeared almost black. Another bird with it was slightly paler and similar to other birds, which had appeared in November. All of these birds seemed to be slightly larger, darker on the upperparts, but not darker on the underparts, than the typical race. They also kept to the fields near the houses rather than in the shrubs where a typical bird had been caught in October. On 17th-18th November a similar situation was found in the Ballyness Bay and Dunfanaghy districts. All of these birds were certainly darker than those, which I am accustomed to seeing in southeast England. It seems possible that in view of the differences, which exist between typical ericetorum [T.p.clarkei], and typical hebridensis that these birds belong to some intermediate population, possibly to the one described by Clancey (1938) as T.ericetorum catherinae [T.p.catherinae … now no longer considered a valid subspecies] and which he located as W.Scotland and the Inner Hebrides.”

Clancey, P.A. (1938). Some remarks on Western Scottish Birds. Ibis 2: 746–754


I owe enormous thanks to Alex Lees, Martin Gray, Neville McKee and Nick Watmough for technical feedback and new insights, all of which I have woven into the text. Julian Hough and Julian Wyllie improved the content in various ways. Angela Ross, curator of natural history at the Ulster Museum, was most helpful and allowed me to examine dozens of Song Thrush specimens. Stuart Rivers provided detailed ornithological directions for my visit to Barra. At the British Trust for Ornithology, Jez Blackburn and Jacqui Clarke facilitated a search for ringing details of all Song Thrushes trapped on the Outer Hebrides. Philip S Redman got in touch through Roger Riddington (British Birds editor) and kindly forwarded his field notes from Tory Island, Co. Donegal, in 1954: which are included in a postscript.


1 Clarke, W.E. The Song-Thrush of the Outer Hebrides — Turdus musicus hebridensis — a new racial form. Scottish Naturalist (1913, pp.53–55, plate 1).

2 The website of the Stornoway Historical Society ( contains a summary that describes the age and composition of the woodland: ‘King James VI granted ownership of Lewis to the Mackenzies of Kintail in 1610. By about 1680, Lord Seaforth had established his estate house, Seaforth Lodge, on the Gearraidh Chruaidh, an area of rough sheiling [grazing] on the west side of Stornoway harbour … The last male descendant of the Mackenzies of Kintail was Francis, Earl of Seaforth, who died in 1815 and in 1844 the Lewis Estate was sold to James Matheson … who was born in Lairg, Sutherland and co-founded the Jardine-Matheson Company in Canton [China] in 1832. Having made his fortune from the Chinese opium trade, he returned to Scotland and, in 1844, purchased Lewis from the Mackenzie trustees for £190,000. Matheson commissioned the renowned architect Charles Wilson to design his new island residence on the site of the Mackenzie’s Seaforth Lodge. Building work started in 1847 and the £60,000 project took seven years to complete. A further £49,000 was spent on transforming the rough grazing land around the new castle into extensive woods and private gardens. The temperate climate and shelter from the initial planting of hardy species, created ideal growing conditions for a wide range of native and imported species.’

3 Witherby, H.F., Jourdain, F.C.R., Ticehurst, N.F. & Tucker, B.W. 1940. The Handbook of British Birds. Volume II, pp.118–119.

4 McGowan, R.Y., Clugston, D.L. & Forrester, R.W. 2003. Scotland’s endemic subspecies. Scottish Birds 24:18–25.

5 Birding World. February 2004, volume 17, number 2, pp 71–75.

6 Forrester, R., Andrews, I., McInerney, C., Murray, R., McGowan, B., Zonfrillo, B., Betts, M., Jardine, D. & Grundy, D. 2007. The Birds of Scotland. Scottish Ornithologists’ Club.

7 Baptista, L. & Welty, J.C. 1988. The Life of Birds. Fourth edition. Thomson (Brooks/Cole).

8 Johnston, R.F. & Selander, R.K. 1971. Evolution in the House Sparrow. Evolution 25:1–28.

9 Burtt, E. H. Jr., Ichida, J. 2004. Gloger’s Rule, feather-degrading bacteria and color variation among Song Sparrows. The Condor 106 (3):681–686. doi:10.1650/7383. ISSN 0010–5422.

10 Gloger, C.W.L. 1833. ‘5. Abanderungsweise der einzelen, einer Veranderung durch das Klima unterworfenen Farben.’ Das Abandern der Vogel durch Einfluss des Climas (The Evolution of Birds Through the Impact of Climate). Breslau: August Schulz, pp.11–24. ISBN 978–3–8364–2744–9. OCLC 166097356.

11 Pallas, P.S. 1811. Zoographia Rosso-Asiatica.

One clap, two clap, three clap, forty?

By clapping more or less, you can signal to us which stories really stand out.