Oxynoemacheilus nasreddini, Yoğurtçuoğlu & Kaya & Freyhof, 2021

Yoğurtçuoğlu, Baran, Kaya, Cüneyt & Freyhof, Jörg, 2021, Oxynoemacheilus nasreddini, a new nemacheilid loach from Central Anatolia (Teleostei: Nemacheilidae), Zootaxa 4974 (1), pp. 135-150 : 138-146

Oxynoemacheilus nasreddini , new species

( Figs. 2–5 )

Holotype. FFR 15588, 54 mm SL; Turkey: Afyon prov.: stream Aksu at Ayvalı, 6 km north of Sincanlı , 38.8101 30.2560.

Paratypes. FFR 15589, 8, 42–57 mm SL; FSJF 3205 , 5 , 57–73 mm SL, same data as holotype .

Material used in molecular genetic analysis. FSJF-DNA 1518 Turkey: Afyon prov.: Aksu stream at Ayvalı village, 6 km north of Sincanlı , 38.81012N 30.25601E (GenBank accession numbers: KJ 554031 , KJ553853 , MW 916396 , MW 916397 , MW 916398 ) .— FSJF-DNA 1658 ; Turkey: Isparta prov: stream Özdere at Eğirler , 38.1976N 31.1074E (GenBank accession numbers: MW 916392 , MW 916393 , MW 916394 , MW 916395 ) .

Additonal material. FFR 15607, 2, 52–53 mm SL; Turkey: Afyonkarahisar prov.: stream flowing into Seyitler reservoir at İscehisar , 38.8160N 30.7870E .— FFR 1537 , 35 , 30–66 mm SL; Turkey: Afyonkarahisar prov.: stream Kali at İnli, 1 km south to Maltepe , 38.6004N 30.8934E .— FFR 1565 , 7 , 54–76 mm SL; Turkey: Konya prov.: stream Deliköyboğazı 1 km south to Ilgın 38.2370N 31.8834E .— FSJF 3209 , 8 , 58–78 mm SL; Turkey: Konya prov.: stream Ali (Karacaağaç) at Doğanhisar , 38.1356N 31.6674E .— IUSHM 2021-1429 , 2 , 63–67 mm SL; FSJF 3096 , 3 , 66–71 mm SL; Turkey: Isparta prov.: Özdere stream at Eğirler village 8 km northeast of Gelendost , 38.1976N 31.1074E .— FFR 15523, 10 , 42–68 mm SL; Turkey: Isparta prov.: Özdere stream at Madenli village 6 km northeast of Gelendost , 38.1813N 31.1008E .— FSJF 2471 , 2 , 60–67 mm SL; Turkey: Isparta prov.: lower stream Çayköy at Koysazı bridge, 37.8415N 30.8916E .— FSJF 2516 , 1 , 61 mm SL; Turkey: Isparta prov.: middle stream Çayköy above Kemerköprü water regulator, 37.8376N 30.9008E .

Diagnosis. Oxynoemacheilus nasreddini is distinguished from O. mediterraneus by having an emarginate caudal fin (middle caudal-fin ray 76–91% of length of longest upper caudal-fin ray vs. forked, 65–76), a slender caudal peduncle (caudal-peduncle depth 1.5–2.1 times in length vs. 1.2–1.5), a slowly decreasing body depth between the dorsal- and caudal-fin bases, at least until the adipose crests if present (vs. almost uniform), and the flank blotches being usually disconnected from the saddles on the back, rarely 1–2 flank blotches connected (vs. usually all connected).

Oxynoemacheilus nasreddini occurs together with O. angorae in the Lake Ilgın basin. It is distinguished from this species by having irregularly shaped and set, vertically elongated blotches on the flank, rarely a mottled or marbled pattern (vs. a series of dark-brown midlateral blotches usually fused into a wide, irregular shaped midlateral stripe, rarely a mottled pattern), the tip of the pectoral fin usually reaching to or slightly beyond the pelvic-fin origin in the male (vs. not reaching), a longer pre-pelvic length (51–55% SL vs. 48–51), a smaller distance between the pelvic- and anal-fin origins (20–23% SL vs. 23–28), and 8–10 pores in the preoperculo-mandibular canal (vs. 10–13).

The new species is distinguished from O. eregliensis from the wider Lake Tuz basin by possession of a more slender caudal peduncle (caudal peduncle depth 1.5–2.1 times in its length vs. 1.2–1.5), and an emarginate caudal fin (middle caudal-fin ray 76–91% of length of longest ray in upper caudal-fin lobe vs. almost truncate, 88–98).

It is distinguished from O. atili from Lake Beyşehir basin by having a series of vertically elongated blotches along the lateral midline or slightly below, rarely a mottled pattern (vs. irregularly set and shaped, often round blotches, usually forming a marbled pattern), anal and pelvic fins without brown blotches (vs. present in individuals larger than 50 mm SL), anal-fin origin and anal-fin base without pigmentation (vs. usually with a brown blotch at anal-fin origin, often with brown anal-fin base), a smaller distance between the pelvic- and anal-fin origins (20–23% SL vs. 23–26), and a longer anal fin (anal fin length 16–19% SL vs. 14–16).

Oxynoemacheilus nasreddini is distinguished from O. anatolicus from Lake Burdur basin and the Dalaman drainage by having a more slender caudal peduncle (caudal peduncle depth 1.5–2.1 times in its length vs. 1.3–1.6), the tip of the pectoral fin usually reaching to or slightly beyond the pelvic-fin origin in the male (vs. not reaching), and the dorsal part of the head and the upper part of the cheek with a vermiculate or marbled pattern (vs. mottled).

It is distinguished from O. theophilii , from the Bakırçay drainage ( Turkey) and Lesbos Island ( Greece) by having a more slender caudal peduncle (caudal peduncle depth 1.5–2.1 times in length vs. 1.3–1.6), a slowly decreasing body depth between the dorsal- and caudal-fin bases, at least until the adipose crests, if present (vs. body depth almost uniform), and a marbled, rarely mottled flank pattern (vs. mottled).

Oxynoemacheilus nasreddini is distinguished from O. germencicus , O. mesudae and O. cinicus from the Büyük Menderes drainage by having the tip of the pelvic fin usually not reaching to the anus (vs. reaching), and a prominent inner axial stripe (vs. absent). Oxynoemacheilus germencicus is very widespread in the Büyük Menderes drainage and it is very variable in its colour pattern, which reach from a marbled pattern with large roundish blotches, to a fine mottled pattern (vs. usually a series of vertically elongated blotches along the lateral midline or slightly below in O. nasreddini ). But in both species, mottled individuals occur. Some O. germencicus also have short bars on the flank, which are set along the lateral midline (vs. usually below in O. nasreddini ) but some O. germencicus have the bars largely below the lateral midline and some O. nasreddini have the bars along the lateral midline. There need to be more research on how to distinguish O. germencicus , O. mesudae and O. cinicus , which form one closely related group in the molecular analysis ( Fig. 1 ).

Description. See Figures 2–5 for general appearance and Table 2 for morphometric data. Medium-sized, slender species. Head long, body depth at dorsal-fin origin 1.2–1.4 times in head length. Body deepest and widest at about midline between nape and dorsal-fin origin. Body width greatest at pectoral-fin base. Body depth slowly decreasing between posterior dorsal-fin base, until vertical of anal-fin base. Section of head roundish, flattened on ventral surface, slightly convex in interorbital space, convex on snout. Snout roundish. Caudal peduncle compressed laterally, 1.5–2.1 times longer than deep. No, or a very rudimentary pelvic axillary lobe at base of pelvic fin, fully attached to flank. Pelvic-fin origin below first or second branched dorsal-fin ray. Anal-fin origin at in front of vertical of midline between dorsal and caudal-fin origins. Pectoral fin reaching to approximately 65–85% of distance from pectoral-fin origin to pelvic-fin origin in female, reaching to, or slightly beyond pelvic-fin origin in male. Pelvic fin usually not reaching anus, reaching to genital papillae; reaching vertical of tip of last dorsal-fin ray or slightly anterior to that point. Anus about 60–90% of an eye diameter anterior to anal-fin origin. Anal fin not reaching caudal-fin base. An elevated, short dorsal and ventral adipose crest on caudal peduncle behind vertical of posterior anal-fin base in some individuals, shallow or absent in other individuals. Largest known individual 78 mm SL.

Dorsal fin with 7½ branched rays, outer margin slightly concave. Anal fin with 5½ branched rays, outer margin straight or slightly concave. Pectoral fin with 8–10 branched rays, outer margin straight. Pelvic fin with 5–6 branched rays, outer margin straight or slightly convex. Caudal fin emarginate, shortest middle caudal-fin ray 76– 91% of longest ray of upper caudal-fin lobe. Caudal fin with 9+8 branched rays. Flank and back covered by scales, scales irregularly set on predorsal back, densely set on predorsal flank below lateral line. Chest and belly without scales. Lateral line usually complete, terminating on hypural complex, sometimes interrupted between anal fin base and hypural complex. Anterior nostril opening at end of a low, ovoid, flap-like tube. Posterior tip of anterior nostril overlapping posterior nostril when folded backwards. One central pore and one or two lateral pore on each side of supratemporal head canal, 10–12 pores in infraorbital canal, 7–10 pores in supraorbital canal, and 8–11 pores in mandibular canal. A suborbital flap in male. Mouth small, arched. Lips thick without furrows, lower lip thicker than upper lip. A median interruption in lower lip. Upper lip with a small and shallow median incision. Processus dentiformis narrow and rounded. Lower jaw rounded, without median notch. Barbels moderately long; inner rostral barbel reaching or not reaching base of maxillary barbel, outer reaching to vertical of posterior nare or anterior eye margin. Maxillary barbel reaching to, or rarely slightly exceeding vertical through posterior eye-margin.

Coloration. Body with yellowish or pale-brown background and dark-brown pattern in live and preserved individuals. Dorsal head and upper part of cheek with a vermiculate or marbled pattern, cheek without pattern in a few individuals. Ventral surface of head yellowish without pattern. Flank mottled in few individuals, usually with a marbled pattern of many vertically elongate, irregularly set and shaped blotches, usually much narrower than interspaces. Blotches above lateral midline much larger than below, shape of blotches often interrupted along lateral line, irregularly set bars on caudal peduncle in some individuals. Blotches on flank loosely connected by an inner axial stripe in preserved individuals. Inner-axial stripe prominent on flank behind dorsal-fin base, absent in life. Blotches not extending to middorsal saddles in most individuals, individual blotches extending to saddles in few individuals. Midlateral blotches usually fine and dissociated into a fine marbles or mottled pattern on predorsal part of flank, less dissociated on postdorsal part. Back with 2–4 irregularly set and shaped predorsal saddles, saddles dissociated into a mottled or vermiculated pattern in few individuals, 3–6 irregularly set and shaped saddles behind dorsal-fin base. Several spots, vermiculations or a mottled pattern between blotches on back and flank above lateral midline. Some individuals with faint, narrow, brown slashes between myomeres on flank above lateral midline. An irregularly shaped dark-brown bar or two dark-brown blotches, often connected to each other at caudal-fin base, indistinct in some preserved individuals. Dorsal fins with many, small brown blotches on rays, forming 2–3 narrow bands. Caudal fin with many small brown blotches on rays, usually forming 2–4 distinct bands. All fins yellowish to pale grey. Anal and pelvic fins hyaline in life, yellowish in preserved individuals, without blotches on rays.

Distribution. Oxynoemacheilus nasreddini is found in tributaries of the endorheic Lakes Akşehir, Eber, Eğirdir and Ilgın in Central Anatolia.

Etymology. The species is named after Nasreddin Hodja, an iconic character and wise man who is famous for his funny anecdotes and take-home messages. Nasreddin Hodja is believed to lived and died in 13 th century in Akşehir (Konya). A noun in genitive, indeclinable.

Remarks. Oxynoemacheilus nasreddini is closely related to O. mediterraneus and both species are distinguished by a minimum K2P distance of 1.2% in their mitochondrial COI sequence. Geiger et al. (2014) make clear, that there is a wide range of species recognised by morphological characters with very different and often very low K2P distance in their mitochondrial COI sequence. As there are no agreed universal thresholds for species demarcation in the COI sequence distance ( Geiger et al. 2014; Kunz 2012), we adopt an iterative taxonomic approach ( Yeates et al. 2011) that asks for the formulation of a species hypothesis and its testing by independent methods. Following this approach, populations without clear morphological differences are treated as conspecific if they have K2P distances smaller than 2%. But, in similar way, if there are clear diagnostic morphological characters, then the molecular distance below an a priori defined threshold is ruled out and the entities are treated as separate species (see also the most recent and relevant studies that applied similar approach: Freyhof et al. 2018a, 2018b; Yoğurtçuoğlu & Freyhof 2018; Fagundes et al. 2020; Malabarba et al. 2021). Naturally, we are aware that “clear diagnostic morphological characters” is not well defined and needs experience in the particular species group. However, it is beyond the scope of this study to discuss this issue in detail.

As O. mediterraneus and O. nasreddini can be well distinguished by the characters given above, both are recognised as separate species. On one hand, the superiority of this approach over the traditional single-character approach (here, solely COI distance) comes from adopting the concept of species as evolving, natural and diagnosable units. On the other hand, mere use of barcoding methods has always the potential to ignore phylogenetically young species that originated a relatively a short time ago ( Ferguson 2002, Kunz 2007, Yoğurtçuoğlu & Freyhof 2018). This might be the case for O. nasreddini which could have recently separated from O. mediterraneus by the combination of complex geological events including the rising of the Sultan Mountains between Akarçay and the Lake District (including Eğirdir and Beyşehir), and the relationships between Lake Eğirdir and the two Mediterranean rivers (Aksu and Köprüçay), which are difficult to reconstruct without data on the timeline of geological events.

We examined Oxynoemacheilus from Lake Eğirdir basin and could sequence three individuals from the stream Özdere (FSJF 3096, IUSHM 2021-1429). The stream Özdere is flowing from the south slopes of Sultan Mountains into Lake Eğirdir. These three individuals show a minimum K2P distance of 0.2% to O. mediterraneus . These five individuals as well as 10 individuals from the same locality (FFR 15523) have an emarginate caudal fin (middle caudal-fin ray 80–90% of length of longest ray in upper caudal-fin lobe vs. 76–91 in O. nasreddini , 65–76 in O. mediterraneus ), a slender caudal peduncle (caudal-peduncle depth 1.4–1.9 times in length vs. 1.5–2.1 in O. nasreddini , 1.2–1.5 in O. mediterraneus ), a slowly decreasing body depth between the dorsal- and caudal-fin bases (decreasing in O. nasreddini , not decreasing in O. mediterraneus i.e. the ratio of body depth at posterior of dorsal fin base to the body depth at caudal fin base is equal or almost equal to 1), and the flank blotches disconnected from the saddles on the back (vs. disconnected in O. nasreddini , usually all connected in O. mediterraneus ). We could further examine three individuals from the stream Çayköy (FSJF 2471, FSJF 2516), flowing from the south to Lake Eğirdir. These fish also have an emarginate caudal fin (middle caudal-fin ray 75–83% of length of longest ray in upper caudal-fin lobe vs. 76–91 in O. nasreddini , 65–76 in O. mediterraneus ), a slender caudal peduncle (caudal-peduncle depth 1.8–2.0 times in length vs. 1.5–2.1 in O. nasreddini , 1.2–1.5 in O. mediterraneus ), a slowly decreasing body depth between the dorsal- and caudal-fin bases (decreasing in O. nasreddini , not decreasing in O. mediterraneus ), and the flank blotches disconnected from the saddles on the back (vs. disconnected in O. nasreddini , usually all connected in O. mediterraneus ). See also Fig. 6 for general appearance of Oxynoemacheilus from Lake Eğirdir and Fig. 7 for O. mediterraneus as well as Table 3 for morphometric data of O. mediterraneus .

All Oxynoemacheilus from Lake Eğirdir basin examined are identified as O. nasreddini as they are differentiated from O. mediterraneus by the same set of morphological characters that distinguish O. nasreddini from O. mediterraneus (see above). The discordance between the morphological and molecular characters might be the result of introgressive hybridization between O. mediterraneus and O. nasreddini . This is likely to have happened through a connection of the Aksu with Lake Eğirdir basin. Similar possible mitochondrial introgression has been posited between Seminemacheilus ispartensis from Lakes Eğirdir basin and S. lendlii from Lake Akşehir, Eber, and Ilgın basins ( Yoğurtçuoğlu et al. 2020). Also, genetic introgression was demonstrated in the frog Pelophelax caralitana from the same region occurring in all peripheral basins, including Akşehir, Eber, Eğirdir and Beyşehir ( Akın et al. 2010).

Lake Eğirdir is connected today to the Aksu through a canal connecting it with Lake Kovada which drains then to the Aksu. The Kovada canal connecting both lakes was dug along a natural streambed transporting excess water from Eğirdir to the Aksu. Loaches from the Aksu drainage might have invaded the lake following this stream and hybridized with O. nasreddini in Lake Eğirdir leading to the potential introgression of mtDNA postulated here. In Oxynoemacheilus , such an invasion of a mitochondrial haplotype was recently suggested between O. elsae and O. brandtii in the Aras River in the Caspian Sea basin ( Freyhof et al. 2021a).

The western Taurids, including Lake Eğirdir, Beyşehir, Suğla as well as Aksu, Köprüçay and Manavgat Rivers constitute a highly complex karst region. Several studies have suggested that these three river drainages receive a significant amount of water from huge karst springs fed mainly through the sinkholes at the periphery and bottom of Lake Eğirdir and Beyşehir ( Ekmekçi, 1993; Değirmenci & Günay, 1992). It is postulated that Eğirdir Lake is hydrologically connected by a subsurface conduit flow to Köprüçay River ( Ekmekçi, 1987; UNDP, 1983). The size of the karst conduits may be large enough to form underground rivers by which some organisms including fish might be able to move by means of their drifted eggs/larvae (Mehmet Ekmekçi, pers. comm., 2021). Evidences from fish fauna support these hypotheses as Pseudophoxinus fahrettini from the upper Köprüçay is most similar to P. handlirschi from Lake Eğirdir; Garra klatti , which was found in Lake Eğirdir still survives in the upper Köprüçay, both species being absent from the Aksu River drainage. These species and potentially also the Oxynoemacheilus might have used existing groundwater connections between Lake Eğirdir and the Köprüçay or there might have been existed some paleo surface drainage pathways between Eğirdir Lake basin and Köprüçay River basin.

Figures

FIGURE 1. Maximum likelihood estimation of the phylogenetic relationships based on the mitochondrial DNA barcode region (GTR model, discrete Gamma distribution for rate differences with 5 categories (+G, parameter = 0.2139). Nucleotide positions with less than 98% site coverage were eliminated, resulting in 514 analyzed positions. Numbers of major nodes indicate bootstrap values from 1000 pseudoreplicates from the ML, NJ and MP method. X denotes highest possible support from all three methods (>95%). Branches with less than 50% bootstrap values were deleted. The tree is drawn to scale with branch lengths depicting number of substitutions per site.

FIGURE 2. Oxynoemacheilus nasreddini, FFR 15588, holotype, male, 54 mm SL; Turkey: stream Aksu.

FIGURE 3. Oxynoemacheilus nasreddini, paratypes, from top, FFR 1537, male, 69 mm SL; Turkey: stream Kali FFR 1565, female, 63 mm SL; stream Deliköyboğazı; FFR 15607, male, 52 mm SL; Turkey: stream flowing into Seyitler reservoir.

FIGURE 4. Oxynoemacheilus nasreddini, FFR 15588, holotype, male, 54 mm SL; Turkey: stream Aksu.

FIGURE 5. Oxynoemacheilus nasreddini, from top: Not preserved, about 70 mm SL; Turkey: stream Battal, 38.2356N 31.8803E; not preserved, about 70 mm SL; same place as FSJF 3205, Turkey: stream Aksu; paratype FSJF 3209, 75 mm SL; Turkey: stream Ali.

FIGURE 6. Oxynoemacheilus nasreddini, top: FSJF 3096, 69 mm SL; Turkey: stream Özdere at Eğirdir; bottom: FSJF 2471, 67 mm SL; Turkey: stream Çayköy.

FIGURE 7. Oxynoemacheilus mediterraneus, from top: FSJF 3111, 48 mm SL; Turkey: Köprüçay; FSJF 3029, 45 mm SL; Turkey: stream Çığırgan; FSJF 2314, 61 mm SL; Turkey: Aksu River.

Tables

TABLE 2. Morphometric data of Oxynoemacheilus nasreddini (holotype FFR 15588, paratypes FFR 15589 n=8; and

	holotype	Holotype, paratypes and additional materials
		Mean	Min	Max	SD
Standard length (mm)	53.5		42.3	69.3
In percent of standard length
Head length	24.1	23.7	22.5	25.1	0.7
Body depth at dorsal-fin origin	17.9	18.3	16.7	21.1	1.2
Body width at dorsal-fin origin	14.1	14.1	11.4	16.9	1.6
Predorsal length	50.0	51.4	49.3	54.4	1.4
Postdorsal length	41.5	38.5	35.5	41.5	1.6
Preanal length	74.9	73.1	70.2	74.9	1.1
Prepelvic length	54.8	53.1	51.0	55.4	1.1
Distance between pectoral- and pelvic-fin origins	30.7	28.9	26.8	30.7	1.5
Distance between pelvic- and anal-fin origins	21.1	21.0	19.6	23.3	0.9
Distance between vent and anal-fin origin	3.2	2.7	1.0	3.7	0.6
Depth of caudal peduncle	9.8	10.6	9.6	11.9	0.5
Length of caudal peduncle	19.0	19.4	17.3	21.9	1.2
Dorsal-fin length	22.0	21.0	19.3	22.9	1.0
Anal-fin height	17.4	17.2	16.0	19.2	0.8
Pectoral-fin length	19.8	21.8	18.2	27.8	2.7
Pelvic-fin length	15.8	16.1	14.7	17.9	1.0
In percent of head length
Snout length	42	39.6	33	43	2.7
Eye diameter	17	16.9	15	19	1.3
Postorbital distance	47	47.6	43	51	1.9
Maximum head width	68	66.9	64	72	2.4
Head depth at eye	52	49.5	47	53	2.0
Interorbital width	28	30.2	26	33	1.7
Length of inner rostral barbel	23	23.6	21	26	1.3
Length of outer rostral barbel	31	32.7	26	38	2.9
Length of maxillary barbel	30	31.0	27	38	2.9

TABLE 2. Morphometric data of Oxynoemacheilus nasreddini (holotype FFR 15588, paratypes FFR 15589 n=8; and additional materials FFR 1565, n=6; FFR 1537, n=3; FFR 15607, n=2). The calculations include the holotype.