Salmo kottelati sp. n., is described from Alakır Stream (Mediterranean basin) in Turkey. It is distinguished from other Anatolian Salmo species by a combination of the following characters (none unique to the species): general body colour greenish to silvery in life; 7–9 parr marks along lateral line; four dark bands on flank absent in both sexes; black ocellated spots few, present only on upper part of flank in individuals smaller than 160 mm SL but in larger both males and females black spots numerous and located on back and middle and upper part of flank; red spots few to numerous, scattered on median, and half of lower and upper part of flank; head long (length 29–33% SL in males, 26–32 in females); mouth large (length of mouth gape 13–19% SL in males, 12–15 in females); maxilla long (length 10–13% SL in males, 8–12 in females); 105–113 lateral line scales; 24–29 scale rows between lateral line and dorsal-fin origin, 17–19 scale rows between lateral line and anal-fin origin; 13–15 scales between lateral line and adipose-fin insertion.

Anatolia, fish distribution, taxonomy, Salmo , new species

Salmo trutta Linnaeus, 1758 has long been considered to be a polymorphic species widely distributed throughout Europe and the Middle East reaching in south and south-east to the Atlas Range (Morocco, Algeria) and to the upper Amu-Darya drainage in Afghanistan (Kottelat 1997). Since many years, a number of subspecies of S. trutta or distinct species then assigned to S. trutta have been described. Most authors considered S. trutta to be a very variable species forming three major ecotypes (sea migratory, lacustrine, and riverine or brook).

Some forms or subspecies of S. trutta distributed in Europe and Asia were resurrected to the species level by Kottelat (1997). Later, species status of some North African species was discussed by Delling and Doadrio (2005) and of Balkan ones by Delling (2003, 2011). Kottelat and Freyhof (2007) tentatively recognised 29 species from European waters and mentioned that the status of several populations and nominal species was still not clear.

Tortonese (1955) reviewed the trouts of Anatolia and reported four subspecies of Salmo trutta: S. trutta labrax Pallas, 1814 from the Çoruh River (Black Sea basin), Lake Çıldır in the Kura drainage (Caspian Sea basin), and Uludağ mountains (Marmara Sea basin); S. trutta caspius Kessler, 1877 from the Kura River (Caspian Sea basin); S. trutta macrostigma (Duméril, 1858) from the Çoruh River (Black Sea basin), and the Çatak Stream (the Tigris system) and described S. trutta abanticus Tortonese, 1955, as a new subspecies, from Lake Abant (a closed lake in northwest Anatolia). Distribution area of S. trutta macrostigma was later considered as including, besides the Çoruh River, the Aegean Sea basin, the Marmara Sea basin, the Trace region, the Mediterranean basin, and the Tigris-Euphrates drainage (Geldiay and Kähsbauer 1967; Kelle 1978; Kuru 1975, 2004; Balık 1984; Bardakçı et al. 1994; Küçük 1997). Salmo trutta labrax was commonly reported from the streams and rivers flowing to the southeastern shore of the Black Sea, and S. trutta caspius from the Kura drainage (Aras 1974; Kuru 1975, 2004; Kutrup 1994; Geldiay and Balık 2007, Tabak et al. 2002; Turan 2003). In addition, S. platycephalus Behnke, 1968 was described from the upper Seyhan drainage (Behnke 1968; Kuru 2004; Schöffmann 2004).

Turan et. al (2010) surveyed all rivers and streams draining to the Black Sea coast in Anatolia, and recognised three morphologically distinct groups of Salmo populations. These groups were identified by earlier authors as S. trutta labrax (drainages of south-eastern Black Sea coast), S. trutta macrostigma (Çoruh River drainage), and S. trutta abanticus (the outlet of the Abant Lake in the Bolu Province). Turan et al. (2010) compared the samples from the Black Sea coast of Anatolia with a sample of true Salmo labrax from the Ulu-Uzen River in Crimea (geographically close to one of the sites of the type locality of S. labrax, Biyuk-ozen River in Kacha River upper reaches in Crimea) and a sample from the Khosta River in Krasnodar Krai in Russia (north-eastern Black Sea coast). The comparison revealed that the Salmo populations from the Black Sea coast of Anatolia are different from S. labrax from the Crimea Peninsula and the Caucasian coast in Russia. Salmo trutta abanticus was evaluated at species level; the two other populations were asserted morphologically distinct, diagnosable, and described as new species: S. coruhensis Turan, Kottelat & Engin, 2010 and S. rizeensis Turan, Kottelat & Engin, 2010. They also suggested that both species in several streams are present in sympatry, although rarely in syntopy. An opinion was proposed by Turan et al. (2010) that the trouts of the Black Sea basin in northeastern Turkey may represent distinct species as upstream resident trouts of different river drainages are genetically closer to each other than to migratory trouts in respective drainages. Tortonese (1955) reported both S. trutta labrax and S. trutta macrostigma from the Çoruh drainage. Turan et al. (2010) concluded that these apparently correspond to S. coruhensis and S. rizeensis respectively. The two species were also recorded from other streams and rivers draining to south-eastern Black Sea coast (see Turan et al. 2010: 336) (Fig. 1). Salmo labrax is distributed in the northern drainages of the Black Sea coast (undoubtly from the northwest Caucasia in Russia to the Danube River) but the exact border between the ranges of S. labrax and S. coruhensis is still not known and the trout from rivers in Georgia still needs a taxonomic study. As to S. coruhensis, it is distributed in the drainages of south-eastern Black Sea coast from the Çoruh River in the north to the Kızılırmak River.

Figure 1. 

Salmo kottelati; Turkey: Antalya Province: Alakır stream a FFR 03180, holotype, 205 mm SL, male b FFR 03181, paratype, 207 mm SL, male c FFR 03181, paratype, 208 mm SL, female d FFR 03181, paratype, 175 mm SL, female e FFR 03181, paratype, 98 mm SL, Juvenile.

Distribution of S. macrostigma is restricted to Algeria (Kottelat 1997; Delling and Doadrio 2005). The other peri-Mediterranean populations referred to as S. macrostigma belong to several species (e.g. S. cettii in Italy and S. farioides in the eastern Adriatic drainages) (Kottelat 1997). In Turkey, besides S. rizeensis, some populations earlier misidentified as S. trutta macrostigma were described as S. tigridis Turan, Kottelat & Bektaş, 2011 (Tigris-Euphrates drainage, Persian Gulf basin), S. labecula Turan, Kottelat & Engin, 2012 (Ecemiş Stream in the lower Seyhan River, east Mediterranean basin), Kartoz and Zindan streams (tributaries of Köprü Stream, a Mediterranean coastal drainage), S. opimus Turan, Kottelat & Engin, 2012 (Tekir, Fırnız and Göçüksu streams in lower reaches of the Ceyhan River) and Alara Stream (a Mediterranean coastal drainage), S. chilo Turan, Kottelat & Engin, 2012 (Akdere Stream in the upper Ceyhan River), S. okumusi Turan, Kottelat & Engin, 2014 (western Euphrates drainage), and S. euphrataeus Turan, Kottelat & Engin, 2014 (northeastern Euphrates drainage).

The taxonomic status of Salmo populations found in the southern Marmara Sea coast and the Trace Region will be discuss in forthcoming papers.

The present paper reports our data on the identity of the resident trout inhabiting Alakır, a small stream draining to the Mediterranean Sea. We conclude that it is an unnamed species. Here, it is described as a new species Salmo kottelati.

Fish were caught with pulsed DC electro fishing equipment. Material is deposited in: FFR, Zoology Museum of the Faculty of Fisheries, Recep Tayyip Erdoğan University, Rize, and CMK, the collection of Maurice Kottelat, Cornol. Measurements and counts were all obtained on wild caught specimens, well preserved, in a straight position. Specimens not fixed straight or damaged were excluded. Most samples include both sexes, juveniles and mature specimens. Most Salmo populations are small, geographically restricted and under great threat because of overfishing and habitat destruction, and it is not advisable to collect and preserve large series of individuals. Colour pattern and variation in shape were observed in the field on additional individuals which were not preserved.

Measurements were taken with digital calipers (0.1 mm accuracy). Counts and measurements follow Hubbs and Lagler (1947), except as follows. Head depth 1: through eye; head depth 2: head depth at occiput; body depth1: body depth at dorsal-fin origin; body depth 2: body depth at anal-fin origin; body width: body width at level of anal-fin origin; adipose-fin height: measured at middle of fin base; length of adipose-fin base: measured from origin to insertion; length of caudal peduncle: measured from anal fin inseriton to middle of caudal-fin base; distance between adipose fin and caudal fin: measured from adipose-fin insertion to middle of caudal-fin base; length of maxilla: from anterior end to posterior end of upper margin of exposed part; length of mouth gape: maximum depth of combined maxilla and supramaxilla; width of mouth gape: measured between corners of mouth gape; length of mouth gape: distance from tip of snout to corner of mouth gape. Lateral-line scales were counted until the posterior extremity of the hypural complex. Scale rows between the adipose fin and the lateral line were counted at the adipose fin insertion. The last two branched dorsal and anal fin rays articulating on a single pterygiophore were counted as 1½. Vertebrae counts were obtained from radiographs and were counted separately as abdominal and caudal vertebrae. Abdominal vertebrae were counted from the first vertebra. The first caudal vertebra is that with its haemal spine fully developed. The count of caudal vertebrae includes the hypural complex. Sex was determined by examination of the gonads. In the description of colour pattern, we use bands to refer to the broad blackish vertical marks on the body, typically positioned behind the gill opening, below the dorsal fin, above the anal fin, and on the caudal peduncle. In some species these bands are known in well preserved specimens only, or in stressed individuals only, and in other species they are visible in situ in undisturbed individuals. A spot is called ocellated when surrounded by a white or very pale ring.

The morphometric and meristic data for S. coruhensis, S. rizeensis, S. abanticus, S. caspius, S. tigridis, S. platycephalus, S. labecula, S. opimus, S. chilo, S. okumusi, and S. euphrataeus are from Turan et al. (2010, 2011, 2012, 2014).

Salmo kottelati is immediately distinguished from all other species of Salmo in Turkey and S. labrax (from the northern Black Sea basin) by having fewer parr marks on the flank (7–9, vs. 10–13). It is further distinguished from S. platycephalus, S. opimus, S. chilo, and S. labecula by the absence of four dark bands on the flank in males and females (vs. presence).

Besides the above differences, S. kottelati is distinguished from S. platycephalus by the presence of red spots on the flank in individuals of all sizes (vs. absence in specimens larger than about 70 mm SL) and the presence of black spots in individuals of all sizes (vs. absence in specimens larger than about 170 mm SL). The new species has fewer gill rakers on the outer side of the first gill arch (18–20, vs. 23–25), a longer head in males (29–33% SL vs. 27–29), a deeper head (depth through eye 14–17% SL in males, 12–15 in females, vs. 12–13 in males, 11–12 in females), and a longer maxilla (10–13% SL in males, 8–12 in females, vs. 8–10 in males, 7–8 in females).

Besides the differences listed above, S. kottelati differs from S. opimus in having fewer lateral-line scales (105–113, vs. 112–120) and by the number and location of the black and red spots on the body in males and in females larger than about 160 mm SL. In S. kottelati, the black spots are numerous in males larger than about 160 mm SL and in females between about 160–190 mm SL, present on median, and half of upper and lower of part flank, and the number of both red and black spots increases with increasing size and age in males. In S. opimus, black spots are few (less than 50), present on back and upper part of flank, and their number does not increase with increasing size and age in males. There are also other differences between males of S. kottelati and S. opimus: males of S. kottelati have a deeper head (head depth through eye 14–17% SL, vs. 12–13), a longer maxilla (10–13% SL, vs. 9–10), and a greater mouth gape (length of gape 13–19% SL, mean 16, vs. 11–14, mean 13).

In addition to the differences mentioned above, S. kottelati is distinguished from S. chilo by the following characters: a slightly convex dorsal profile of head (vs. strongly convex), a pointed snout (vs. blunt), a slightly subterminal mouth (vs. conspicuously subterminal), non-flesh maxilla and lower lip (vs. flesh), length of the maxilla markedly sexually dimorphic (the length in males markedly longer than that in females, vs. not sexually dimorphic); black spots on the body roundish (vs. irregularly shaped). The new species also differs from S. chilo by having more numerous scale rows between the lateral line and the anal-fin origin (17–19, vs. 15–17). When compared to males of S. chilo, males of S. kottelati have a deeper head (depth through eye 14–17% SL, vs. 12–13), and a longer maxilla (length 10–13% SL, mean 12, vs. 9–10, mean 9).

Besides the differences mentioned above, S. kottelati differs from S. labecula by the presence of red spots on the flank (vs. absence in specimens larger than about 70 mm SL) and the black spots on body roundish (vs. irregularly shaped). Salmo kottelati has more numerous scale rows between the lateral line and the anal-fin origin (17–19, vs. 16–17) and fewer gill rakers (18–20, vs. 21–23). It also differs from S. labecula by having a greater predorsal distance (49–52% SL, mean 50 in males, 47–51, mean 49 in females, vs. 45–48% SL, mean 46 in males, 44–47, mean 45 in females). Moreover, S. kottelati has a longer head in males (29–33% SL, mean 31, vs. 27–29, mean 28) and a deeper head in males (depth through eye 14–17% SL, mean 15, vs. 11–14, mean 13).

Salmo kottelati is further distinguished from S. tigridis by the number and locations of black spots on the body in specimens larger than about 160 mm SL. In S. kottelati, black spots are numerous in males larger than about 160 mm SL and in females between about 160–190 mm SL, present on median and half of upper and lower part of flank, and the number of both red and black spots increases with increasing size and age in males. In S. tigridis, black spots are few (less than 50), present on back and the upper part of flank and their number does not increase with increasing size and age. The new species also differs from S. tigridis by having fewer scale rows between the lateral line and the dorsal-fin origin (24–29, vs. 32–35), fewer scale rows between the lateral line and the anal-fin origin (17–19, vs. 22–26), and fewer scale rows between the adipose fin insertion and the lateral line (13–15, vs. 19–20). Besides the above listed differences, S. kottelati males are distinguished from S. tigridis males by a longer (29–33% SL, mean 31, vs. 25–28, mean 27) and deeper head (depth at nape 18–20% SL, mean 19, vs. 17–18, mean 17), a longer maxilla (10–13% SL, mean 12, vs. 8–9, mean 9), and a wider mouth gape (11–13% SL, mean 12, vs. 9–10, mean 10).

Salmo kottelati is further distinguished from S. abanticus by the presence of red spots on the body in specimens larger than about 210 mm SL (vs. absent), the shape of the black spots on the flank (round, vs. polygonal), the shape of the ring around black spots (circular, vs. polygonal), and the size of the black spots (about equal to the pupil, vs. markedly larger than the pupil). It has fewer lateral line scales than S. abanticus (105–113, vs. 113–121). Besides the differences mentioned above, it also differs from S. abanticus in having a longer head (29–33% SL, mean 31 in males, 26–32, mean 29 in females, vs. 26–29, mean 27 in males, 24–26, mean 26 in females). Males of S. kottelati differ from males of S. abanticus in having a greater predorsal distance (49–52% SL, vs. 47–48), a longer maxilla (10–13% SL, mean 12, vs. 9–10, mean 10), a greater eye diameter (7–9% SL, mean 8, vs. 6–7, mean 6), and a deeper head at nape (18–20% SL, mean 19, vs. 17–19, mean 17).

Salmo kottelati is further distinguished from S. coruhensis by fewer scale rows between the anal-fin origin and the lateral line (17–19, vs. 19–23), fewer scale rows between the adipose fin insertion and the lateral line (13–15, vs. 15–17), a longer (29–33% SL, mean 30.9, vs. 26–28, mean 27.3) and deeper head in males (depth through eye 14–17% SL, mean 15, vs. 12–14, mean 13).

Salmo kottelati is further distinguished from S. rizeensis by the number and location of black spots on the body. In S. kottelati, black spots are numerous (60 and more) in males larger than about 160 mm SL and in females between about 160–190 mm SL, present on median, and half of upper and lower of part flank, and the number of both red and black spots increases with increasing size and age in males. In S. rizeensis, black spots are few (less than 40), present on back and upper part of flank, and their number does not increase with increasing size and age. The new species also differs from S. rizeensis by the general body colour (greenish to silvery in life, vs. brownish), the adipose fin almost reaching the base of the caudal fin in males larger than 200 mm SL (vs. not reaching in all-sized specimens), and fewer lateral-line scales (105–113, vs. 114–120).

Salmo kottelati is further distinguished from S. caspius from the upper Kura drainage by having fewer lateral line scales (105–113, vs. 112–119), fewer scale rows between the lateral line and the anal-fin origin (17–19, vs. 19–22) and fewer scale rows between the insertion of the adipose fin and the lateral line (13–15, vs. 15–17). In Salmo kottelati, the general body colour is greenish to silvery in life (vs. brownish). Morover, males of S. kottelati have a deeper head than that of S. caspius (head depth at nape 18–20% SL, vs. 17–18).

Salmo kottelati is further distinguished from S. okumusi by the absence of four dark bands on the flank in males and females (vs. very faintly marked or indistinct in small specimens but distinct in specimens larger than about 230 mm SL); the parr marks vertically oblong (vs. vertically elongate); the black spots circular (vs. irregularly shaped), bigger black spots on dorsal fin (slightly smaller than pupil, vs. smaller than half pupil). Salmo kottelati has a longer head than S. okumusi (29–33% SL in males, 26–32 in females, vs. 26–27 in males, 25–26 in females). Males of S. kottelati differs from males of S. okumusi by having a longer maxilla (10–13% SL, vs. 9–10) and a deeper head (at nape 18–20% SL, vs. 16–17).

Salmo kottelati is further distinguished from S. euphrataeus by the general body colour and the number and position of the black spots on the body: in S. kottelati, black spots are commonly numerous in males and increasing the number with size (vs. few and not increasing the number) and located on median, and half of upper and lower parts of the flank (vs. restricted to the upper part of the flank, mostly in its anterior area). The new species also differs from S. euphrataeus in having fewer lateral-line scales (105–113, vs. 112–120), fewer scale rows between the dorsal-fin origin and the lateral line (24–29, mean 26, vs. 28–31, mean 29), and fewer scale rows between the anal-fin origin and the lateral line (17–19, vs. 19–23).

Salmo kottelati differs from resident S. labrax by a large black spot behind the head (larger than the pupil but smaller than the eye, vs. equal or smaller than the pupil), black spots on the body smaller than the pupil (vs. larger), more numerous red spots on the body (red spots few to numerous, scattered on median, and half of lower and upper part of the flank, vs. few, one or two irregularly rows of spots, scattered along the lateral line or, sometimes, below it). Salmo kottelati has fewer lateral line scales (105–113, vs. 112–121), fewer scale rows between the dorsal-fin origin and the lateral line (24–29, vs. 28–32), fewer scale rows between the lateral line and the anal-fin origin (17–19, vs. 19–23), fewer scale rows between the adipose fin insertion and the lateral-line (13–15, vs. 15–16), and more numerous gill rakers (18–20, vs. 16–18). Males of S. kottelati are further distinguished from males of S. labrax by having a longer head (29–33% SL, vs. 25–28), a greater predorsal length in males (49–52% SL, vs. 46–47), a deeper head (depth through eye 14–17% SL, vs. 11–13), a longer maxilla (10–13% SL, vs. 9–10), a wider mouth gape (11–13% SL, vs. 8–10), and a longer mouth gape (13–19% SL, mean 16, vs. 12–14, mean 13).

Salmo kottelati can be further distinguished from S. cf. coruhensis from the Gönen Stream (southern Marmara Sea) by fewer lateral-line scales (105–113, vs. 115–121), fewer scale rows between the lateral line and the anal-fin origin (17–19, vs. 20–23), fewer scales between the adipose fin insertion and the lateral line (13–15, vs. 15–17), a larger adipose fin (depth 7–11% SL, vs. 4–7), and a deeper caudal peduncle (depth 10–13% SL, vs. 8–10). Males of S. kottelati differs from males of S. cf. coruhensis by a longer head (29–33% SL, vs. 24–28), a greater predorsal distance (49–52% SL, 45–47), a longer mouth gape (13–19% SL, mean 16, vs. 11–14, mean 12), a narrower mouth gape (width 11–13% SL, vs. 8–10), a longer maxilla (10–13% SL, vs. 8–10), and a deeper head (depth through eye 14–17% SL, vs. 11–14).

Salmo kottelati is further distinguished from S. cf. rizeensis from the Rezova Stream (Trace Region) by fewer lateral-line scales (105–113, vs. 114–121), fewer scale rows between the dorsal-fin origin and the lateral line (24–29, vs. 29–34), fewer scale rows between the lateral line and anal-fin origin (17–19, vs. 20–23), fewer scales between the adipose fin insertion and the lateral line (13–15, vs. 16–17), a larger adipose fin (depth 7–11% SL, vs. 5–7), and a deeper caudal peduncle (depth 10–13% SL, vs. 9–10). Males of S. kottelati differs from males of S. cf. rizeensis by a longer (29–33% SL, vs. 26–29) and deeper head (depth through eye 14–17% SL, vs. 12–14).

We owe thanks Maurice Kottelat, the reviewers and Nina G. Bogutskaya for comments on earlier variants of the manuscript, and also to Sema Turan for assistance in laboratory studies. This work was supported by The Scientific and Technical Research Council of Turkey (Project No. TOVAG–106 O 259).