Arctic charr in Britain and Ireland – 15 species or one?


C. E. Adams, University Field Station, Institute of Biomedical and Life Sciences, University of Glasgow, Glasgow G63 0AW, UK;


Abstract –  From both a modern and a historical perspective there is little doubt that the Arctic charr, Salvelinus alpinus, in Britain and Ireland (as well as elsewhere) is a ‘difficult’ species. Historically 15 separate species have been recognised from populations in Britain and Ireland and there have been recent attempts to reassert these specific names. Here we review the evidence for the status of these ‘species’. We conclude that the evidence for these 15 being afforded full species status is poor. However, both historical and contemporary data show that Salvelinus alpinus (Linnaeus 1758) in Britain and Ireland exhibits levels of variability in form that is much greater than in other species. We argue that a comprehensive genotypic and phenotypic survey of charr populations in Britain and Ireland is required to determine the full extent of variability and the status of populations with a view of providing suitable protection.


The concept of ‘the species’ has long been a controversial area of biology (see for example, Mayr 1982). Universal acceptance of a single, clear definition of a species has been elusive (Eldredge 1995) and, as a result, there may be as many as 22 variant concepts of species in current usage (Mayden 1997).

Discussion of the theoretical and philosophical concepts defining a species is well outwith the scope of this paper (but see Mayr 1988 for entry to this field), however, important to note here is that some groups of organisms have proven particularly taxonomically difficult to assign to species under the concepts that can be used to define species. Organisms that reproduce asexually for example, have challenged taxonomists because of the difficulty of defining populations which are reproductively isolated from one another (Embley & Stackebrandt 1997). For organisms subject to domestication, concepts of evolutionary units as species are redundant because evolutionary change is driven by human selection (Hawkes 1997).

Similarly, populations with a disjunct natural distribution and where there is significant phenotypic variation between populations challenges the allocation and definition of species within these populations (Kottelat 1995).

One fish that displays both a disjunct distribution and a high degree of phenotypic variation is the Arctic charr S. alpinus. The Arctic charr is a freshwater fish of the northern hemisphere. Although anadromous at northern latitudes, below 65°N it completes its life cycle without a sea-going phase (Johnson 1980). Globally there are thought to be at least 50,000 separate populations (Maitland 1995). The Arctic charr displays a very high degree of variance in phenotype over a wide range of characteristics – morphological, meristic, behavioural and in life history (see Johnson 1980; Klemetsen et al. 2003). This variation takes the form of distinct phenotypic variation between populations inhabiting unconnected freshwater systems (Adams et al. 2006a). However, systems supporting sympatric polymorphic groups exhibiting distinct ecological, morphological and life-history traits which, in at least some places, exist as noninterbreeding units, are also common across its range (Gíslason et al. 1999; Klemetsen et al. 2003; Adams et al. 2006b).

This pattern of variation has led to genuinely held and contradictory views on the status of the arctic charr, S. alpinus as a single species. Behnke (1972, 1989) argues that European charr are monophyletic having derived from a relatively recent common ancestor and are a single subspecies Salvelinus alpinus alpinus whose differentiation has occurred during the past 60,000–70,000 years. The only exception to this being the ‘teifseeaibling’ of Bodensee, which Behnke (1972) regards as a full species, Salvelinus profundus.

Similarly, Savvaitova (1995) argues that charr morphotypes should be regarded as ‘nontaxonomic adaptive types’ manifesting phenotypic plasticity in response to environmental variability. Without this approach, argues Savvaitova (1995), it would be possible to define a new species for every waterbody containing charr.

In contrast, some taxonomists consider that there is a logical taxonomic basis for subdividing the arctic charr. In Russia for example, Viktorovskii (1978) and Glubokovskii & Chereshnev (1981) have argued that there are up to 23 separate species (Savvaitova 1995).

Recently, Kottelat (1995, 1997) has challenged the single species views of Behnke and Savvaitova, arguing for a pragmatic approach to the taxonomic difficulties posed by charr: ‘To me it does not make sense to recognise different stocks as biological species and to manage them as species, but not to call them species; the coherent way to handle the problem is that an entity recognised as species using one or the other species concept be called species and be handled as such, in systematics, ecology and resources management.’ Expression of such views have led to calls for a re-evaluation of the taxonomic status of the Arctic charr.

Although for practical purposes, most modern ichthyologists accept a single species in Europe (S. alpinus) (Fig. 1), historically, 15 additional species of charr have been described from waters in Britain and Ireland; five by Gunther (1862, 1863, 1865), one by Maitland (1881) and nine by Regan (1908, 1909a,b) (Table 1). Kottelat (1997) provisionally accepted these 15 plus one further, Salvelinus youngeri (previously described as a subspecies; Friend 1956) but in a cautionary note points out that the status of these putative species needs re-examining and if independent lineages can be established, then they should be accepted as valid species. To add further to the confusion, Wheeler et al. (2004) added 15 of Kottelat's (1997) tentative species (but not S. youngeri) to the widely used list of the common and scientific names of fishes of the British Isles (Wheeler 1992) before the taxonomic review called for by Kottelat (1997) has taken place.

Figure 1.

 A ‘typical’ charr? A drawing of a plankton feeding charr from Loch Doon, Scotland.

Table 1.   Comparative morphological data on the 15 species of charr originally described from Britain and Ireland.
  1. 1, greatest depth of body as a % of length; 2, position of mouth (ter, terminal; sup, superior; inf, inferior); 3, pectoral fin as a % of head length; 4, dorsal fin rays; 5, anal fin rays; 6, dentition (str, strong; mod, moderate; fee, feeble); 7, branchiostegals; 8, gill rakers; 9, lateral scales; 10, vertebrae.


Here, although we do not undertake the taxonomic review of the charr of Britain and Ireland which is clearly required, we do examine the criteria used to define the putative species from this region and the historical circumstances under which they were defined, with the aim of examining their taxonomic validity.

Species described from Britain and Ireland

For each nominal species, the most critical feature observed by the author is given. A summary of morphological and meristic characteristics used in the definition of the species is presented in Table 1 and the number of individual fish used to derive these observations is presented in Table 2.

Table 2.   Numbers of specimens used in early descriptions of Salvelinus species.
SpeciesNumber of specimens
coliimany (Gunther 1863)
fimbriatus1 female (Regan 1908)
gracillimus4 (Regan 1909a)
grayi2 m (Gunther 1862)
inframundus2 (Regan 1909a)
killinensis6 (Gunther 1865)
lonsdalii2 (Regan 1909a)
mallochi4 (Regan 1909a)
maxillaris11 (Regan 1909a)
obtusus14 (Regan 1908)
perisii<20 all male (Gunther 1862)
scharffi1(Regan 1908)
struanensis>20 (Maitland 1881)
trevelyani1 m (Regan 1908)
willoughbii2 m (Gunther 1862)

Salvelinus colii, Cole's charr (Fig. 2), described from Lough Eske by Gunther (1862). Recorded also from Loughs Derg, Gortglass, Currane, Conn (now extinct), Mask, Corrib (now extinct), Inagh, Ballinahinch, Glendalough, Derryneen, Glenicmurrin, Kylemore, Glendawough, Sessiagh, Iskanamacteery, Cloonee and Ennistimon

Figure 2.

Salvelinus colii, Cole's charr (Gordon c. 1902).

‘… the body usually more compressed and deeper (depth 3 1/3 to four inches in the length), the caudal peduncle shorter and deeper (its least depth nearly one half the length of the head), the pectoral fins usually longer …’ (Regan 1911).

Salvelinus fimbriatus, Coomasaharn charr (Fig. 3a,b), described from Lough Coomasaharn by Regan (1908). No other recorded sites.

Figure 3.

Salvelinus fimbriatus, Coomasaharn charr (photo courtesy of F. Igoe; Regan 1908).

‘… takes its specific name (fimbriatus, fringed) from the long and numerous gill rakers.’ (Regan 1911).

Salvelinus gracillimus, Shetland charr (Fig. 4a,b), described from Loch Girlsta by Regan (1909a). There are no other recorded sites and this is the only population of charr on Shetland.

Figure 4.

Salvelinus gracillimus (Shetland charr) (Regan 1909a, 1911.

‘This species has the body more elongate than any other Char, the greatest depth being contained from five and a half to six and a half times in the length of the fish, measured to the base of the caudal fin.’ (Regan 1911).

Salvelinus grayi, Gray's charr (Fig. 5), described from Lough Melvin by Gunther (1862). No other recorded sites.

Figure 5.

Salvelinus grayi, Gray's charr (Regan 1911).

‘… the body usually more compressed and deeper (depth 3 1/3 to 4 inches in the length), the caudal peduncle shorter and deeper …’Regan (1911). ‘The scales are very conspicuous, comparatively much larger than in any other British species.’ (Gunther 1862).

Salvelinus inframundus, Orkney charr (Fig. 6a,b), described from Loch Heldal by Regan (1909a), where it is now extinct. There are no other recorded sites and this was the only population of charr on Orkney.

Figure 6.

Salvelinus inframundus, Orkney charr, now extinct (Regan 1909a, 1911).

‘… a well-defined species … the head is short, the eye small, the interorbital region convex and rather broad, and the fins small.’Regan (1911).

Salvelinus killinensis, Haddy (Fig. 7a,b), described from Loch Killin by Gunther (1865). A similar form recorded from Loch Roy.

Figure 7.

Salvelinus killinensis, Haddy (Regan 1911 and Gunther 1865).

‘… a most remarkable form, which, in the excessive development of its fins, differs from all the other species known to me from Great Britain and the continent of Europe.’ (Gunther 1865).

Salvelinus lonsdalii, Lonsdale's charr (Fig. 8a,b), described from Haweswater by Regan (1909a). Recorded also from Ullswater where it is now extinct.

Figure 8.

Salvelinus lonsdalii, Lonsdale's charr (Regan 1909a, 1911).

‘Especially noteworthy, however, is the great length of the lower jaw …’ (Regan 1911).

Salvelinus mallochi, Malloch's charr (Fig. 9a,b), described from ‘Loch Scourie’ (presumed to be Loch a'Bhadaidh Daraich) by Regan (1909a). There are no other recorded sites.

Figure 9.

Salvelinus mallochi, Malloch's charr (Regan 1909a, 1911).

‘A rather short-headed, blunt-snouted, and small-mouthed Char, with small scales…’ (Regan 1911).

Salvelinus maxillaris, Large-mouthed charr (Fig. 10a,b), described from a lochan on Ben Hope by Regan (1909a). A similar form recorded from Loch Stack.

Figure 10.

Salvelinus maxillaris, Large-mouthed charr (Regan 1911).

Named ‘… on account of the notable length of the maxillary, which extends back far beyond the eye in adult males;’ (Regan 1911).

Salvelinus obtusus, Blunt-snouted charr (Fig. 11a,b), described from Lough Luggala by Regan (1908). Also recorded from Loughs Dan (now extinct), Leane, Muckross and Acoose.

Figure 11.

Salvelinus obtusus, Blunt-snouted charr (Regan 1908, 1911).

‘It is distinguished from other Irish Char by the short, blunt snout and the rounded lower jaw which is included within the upper when the mouth is closed.’Regan (1911).

Salvelinus perisii, Torgoch (Fig. 12), described from Llyn Peris by Gunther (1862), where it is now extinct. Recorded also from Llyn Padarn, Llyn Cwellyn and Llyn Bodlyn.

Figure 12.

Salvelinus perisii, Torgoch (Regan 1911).

‘The head is long, with the snout somewhat produced and pointed, the mouth large and oblique, the lower jaw long …’ (Regan 1911).

Salvelinus scharffi, Scharff's charr (Fig. 13), described from Lough Owel by Regan (1909b) where it is now extinct. Recorded also from Lough Ennell.

Figure 13.

Salvelinus scharffi, Scharff's charr (Regan 1911).

‘… the snout is more acutely conical, the mouth smaller and more oblique, etc., and in the smaller scales, which number 186 in the longitudinal series.’ (Regan 1911).

Salvelinus struanensis, Struan charr (Fig. 14a,b), described from Loch Rannoch by Maitland (1881). A similar form recorded from Loch Leven where it is now extinct.

Figure 14.

Salvelinus struanensis, Struan charr (Regan 1909a, 1911).

Typically has a ‘short blunt snout and inferior mouth; the rounded lower jaw is shorter than the upper and included within it when the mouth is closed; the maxillary is short and broad …’ (Regan 1911).

Salvelinus trevelyani, Trevelyan's charr (Fig. 15a,b), described from Lough Finn by Regan (1908). There are no other recorded sites.

Figure 15.

Salvelinus trevelyani, Trevelyan's charr (Regan 1908, 1911).

‘… longer head, narrower interorbital region, produced pointed snout, and strong dentition.’Regan (1911).

Salvelinus willoughbii, Willoughby's charr (Fig. 16), described from Windermere by Gunther (1862). Recorded also from Coniston Water, Goats Water, Wast Water, Ennerdale Water, Buttermere, Crummock Water, Lowes Water, Loch Grannoch (now extinct), Loch Dungeon (now extinct), Loch Doon, Loch Builg, Loch Bruich, Loch Morie, Loch Borollan, Loyal Baden, Loch Calder and Loch Fada.

Figure 16.

Salvelinus willoughbii, Willoughby's charr (Regan 1911).

‘… there are eight to ten branched rays in the dorsal fin, and seven to nine in the anal; the scales in the longitudinal series vary from 160–200 …’ (Regan 1911).


The definition and discrimination of what is, and what is not, a species is neither an abstract academic activity nor a semantic argument, it has profound practical implications. The number of species is at the heart of concepts of biological diversity, the cornerstone of modern concepts of conservation (Claridge et al. 1997). The species defines the unit used to frame conservation policy and ultimately legislation. It is used in management to define biologically coherent groups subject to exploitation. It is used more than any other unit, to identify a group of organisms subject to common threats and pressures, it is the unit used in surveys to assess conservation status and to assess habitat value or its degradation. It is also used to define strategies and set priorities in conservation and management (Kottelat 1997).

It is clear that the taxonomic status of the species known as the Arctic charr is as unclear now as it was in the 19th century when Hamilton (1843) (quoted in Day 1887) observed that ‘this …… fish is liable to great variation, and this has rendered its synonymy and history somewhat confused.’

Attempts have been made to avoid the difficulties of the taxonomy of the Arctic charr. Savvaitova (1995) for example argues that Arctic charr should be united into a single ‘superspecies’ the S. alpinus complex, thus bringing together groups with differing levels of phenotypic divergence, e.g., morphotypes, subspecies and species together into a single taxonomic grouping. Behnke (1972), however, regards this approach as ‘a semantic cloak for our ignorance’ and there is no doubt that this approach has little practical value for management, conservation and communication of research (Kottelat 1997).

However, the taxonomic position for British and Irish charr suggested by Kottelat (1997) and confirmed by Wheeler et al. (2004) based on species descriptions from the early 20th century is also unsustainable. The argument for full specific status for the 15 putative species examined here is, at best, poor. There are several reasons for this:

  • 1Seven of the species are described from just one or two specimens whereas there is known to be significant variation within charr populations.
  • 2Many of the characters used to describe the putative species are subjective.
  • 3Ichthyologists a century ago had little awareness of sympatric populations of Arctic charr. For example, Loch Rannoch, from which Maitland (1881) described S. struanensis, is now believed to hold three separate populations (Adams et al. 1998).
  • 4A number of the charr populations which were known at the time were only somewhat casually ascribed to a particular species. For example Regan (1911) notes Arctic charr from seven Scottish lochs which ‘have their peculiarities’ but which he was ‘not at present inclined to separate specifically from S. willoughbii.’
  • 5In addition, many new charr lochs have been discovered since the original 15 charr species have been discovered and so it could be assumed that quite a number of new ‘species’ have not yet been described. Less than 100 charr populations were known in Great Britain and Ireland around 1900, whereas by 2000, well over twice that number had been recorded (Maitland 1995).

The Arctic charr situation in this region is not unique. Historically, there have been at least 15 species of Salvelinus described from North America; up to 29 species or subspecies described in Europe; and 12 from Siberia and the Far East (see discussion in Behnke 1972 and Savvaitova 1995).

Both historical and more contemporary evidence (see, e.g., Alexander & Adams 2000; Adams et al. 2006b) does provide the consensus that Arctic charr in Britain and Ireland are highly variable to an extent that differs very significantly from other species and that it is highly likely that under any modern objective criteria for the definition of species, several distinct species would be defined in this region. However, until this taxonomic review takes place there is little evidence to support full specific status for the charr reviewed here.

In reality, there is more objective evidence for full specific status for the sympatric populations which have only been discovered relatively recently (Adams et al. 1998; Fraser et al. 1999; Kettle-White 2001; Adams et al. 2006a). As Kottelat (1997) points out ‘… the data… do not allow a determination of the exact number of involved species …’ Thus we conclude that a taxonomic and populations status review of British and Irish charr is now urgently required.

The urgency for this review is compounded by the fact that we are gradually losing populations (Maitland 1995; Maitland et al. 2006) and the loss of three of the populations of the above type ‘species’ together with many other charr stocks is a matter of considerable concern. What is required is a concerted programme to (i) check the status of all stocks, especially those which have not been examined for many decades; (ii) survey for populations which have not yet been recorded; and (iii) use the material so gathered to define the nature of each stock in both phenotypic and genotypic terms. In our view, such a review is highly likely to lead to the identification of populations that meet the criteria for full species status. Until this is completed, we agree with the sentiments of Seeley (1886)‘… that the constituent varieties of those species should be kept distinct, and treated as though they were species in process of development, still showing their connection with the parent stock.’


Figs. 1 and 4–16 were derived from originals published by Regan (1911); the picture in Fig. 7 was originally published by Gunther 1865. Fig. 1 was drawn by Robin Ade. Fig. 3 was kindly provided by F. Igoe. This work was supported by EC grant QLRI-CT-2001-00007.