Identity of Squalius (Actinopterygii, Cyprinidae) from Istra Peninsula in Croatia (Adriatic Sea basin)

Abstract A chub of previously ambiguous identity from the Boljunscica and Pazincica rivers (south-eastern Istra Peninsula) was studied and compared with geographically close Squalius squalus, Squalius zrmanja, and Squalius janae recently described from the Dragonja River drainage in the Adriatic Sea basin in Slovenia. It was shown that the chub from the south-eastern Istra Peninsula differs from all know species of Squalius but one: Squalius janae. Three samples examined from Boljunscica and Pazincica rivers and Squalius janae from its type locality, Dragonja River, show the following characters typical for the latter species: a long head (the head length 27–32% SL); a pointed conical snout with a clearly projecting upper jaw; a long straight mouth cleft, the lower jaw length (39–45% HL) exceeding the caudal peduncle depth; a large eye; commonly 9? branched anal-fin rays; commonly 44 total vertebrae (24+20 or 25+19); bright silvery colouration, scales easily lost; iris, pectoral, pelvic and anal fin pigmentation with yellow shades. The data on the distribution of Squalius chubs in the northern Adriatic basin support the assumption that the range of Squalius janae is determined by the geology of the Trieste Flysch Basin and the Pazin Flysch Basin forming the base of the Istra Peninsula. The distribution pattern of this species does not support a simple model of fish dispersal and a complete connectivity within the whole Palaeo-Po historical drainage. Indeed, it indicates a disrupted surface palaeohydrography that was heavily fragmented by karstification in the whole Dinaric area.

Th e purpose of the present article is to describe the chub from Pazinčica and Boljunšćica rivers (south-eastern Istra), compare it with S. janae and S. squalus, and decide on its identity.

Material and methods
Measurements were taken point to point to the nearest 0.1 mm following Bogutskaya and Zupančič (1999). Th e standard length (SL) was measured from the tip of the upper jaw to the end of the hypural complex. Th e length of the caudal peduncle was measured from behind the base of the last anal-fi n ray to the end of the hypural complex, at mid-height of caudal-fi n base. Head length (HL) and interorbital width were measured including the skin fold. Postdorsal length was measured from the dorsal-fi n insertion to the posterior end of the hypurals, and the dorso-hypural distance was taken from the origin of the dorsal fi n to the posterior end of the hypurals. A further character was added from Doadrio et al. (2007): a point where the dorso-hypural distance, which is taken from the dorsal-fi n insertion to the end of the hypural complex, falls when measured forward. Th e last two branched rays articulated on a single pterygiophore in dorsal and anal fi ns were counted as "1½". Total lateral line scale counts include all pored scales, from the fi rst one just behind the posttemporal bone to the posteriormost one located on the bases of the caudal-fi n rays. Osteological characters were examined from radiographs.
A map showing the localities of samples from the Istra Peninsula is given in Fig. 1.

Description of specimens from Boljunšćica and Pazinčica rivers.
Morphometric data of 30 specimens from the Boljunšćica and Pazinčica rivers are given in Table 1, and the general appearance can be seen from Figs 2-5. Th e body is elongate, depth at the dorsal-fi n origin is 21-24% SL. Th e head is long, its length, 27-31% SL, is greater than the body depth being 113-131% of the latter, and exceeds the caudal peduncle depth by a factor of 2.6-3.2. Th e upper head profi le is almost straight, the snout is pointed, and the preorbital part of the head is almost triangular in lateral view. Th e mouth is subterminal (Figs 2, 3) or almost terminal though never clearly terminal even in smaller specimens (Fig. 4), and the upper jaw is clearly projecting beyond the lower jaw. Th e mouth cleft is straight, oblique, and the lower jaw-quadrate junction is commonly well visible forming a distinctive obtuse angle (Figs 2-5); the lower jaw-quadrate junction is on the vertical through the middle of the eye. Th e lower jaw length, 39-44% HL, exceeds the caudal peduncle depth by a factor of 1.0-1.3, commonly 1.1-1.2. Th e length of the lower jaw is always greater than both the operculum depth (being 106-116% of the former) and the interorbital width. Th e eye is large, its horizontal diameter being 20-25% in HL, and 54-71% in interorbital width. Th e eye diameter negatively correlates with the fi sh size as it can be seen from comparisons of larger and smaller specimens (Figs 2-3  and 4), for example, it is 26-28% in HL in specimens of 50-60 mm SL and 20% in HL in a specimen of 222 mm SL (NMW 49250) though we did not perform a statistic comparison.     Th e dorsal fi n has 3 simple and 8½ branched rays in all specimens. Its outer margin is straight. Th e dorsal fi n is located slightly behind the end of the pelvic-fi n base. Th e dorso-hypural distance falls when measured anteriorly in the posterior half of the eye, rarely at the posterior eye margin. Th e dorsal fi n is high, its depth being 16-19% in SL. Th e anal fi n has 3 simple and 9½ branched rays; 8½ branched rays were found only in two specimens. Th e anal-fi n outer margin is slightly to markedly convex.
In live specimens, the overall colouration has a strong silvery tint, and the back is only slightly darker than the fl anks and the belly. No conspicuous dark reticulated pattern has been seen in live specimens. Th e iris, anal and pectoral fi n pigmentation has yellow shades, and is never bright (Fig. 2). Formalin fi xed and ethanol stored specimens keep silvery-grey colouration with no brownish or bronze shades; yellow pigments are often lost in preserved specimens. Pigmentation on scales shows a reticulate pattern with comparatively fewer pigment dots along the outer scale margins. A concentration of pigment on scale pockets forming dark vertical spots is commonly visible in larger specimens (Fig. 2) though not pronounced in a small specimen (50 mm SL, Fig. 3).
Th e Pazinčica and Boljunšćica specimens and the historical sample from Čepić Lake examined in this study are thus identifi ed as S. janae based on the diagnostic characters of the latter (Bogutskaya and Zupančič 2010), see Table 1 and Figs 2-5.
Comparative remarks. Th e Pazinčica and Boljunšćica specimens and the Čepić Lake historical sample diff er from S. squalus in general and from specimens from Osapska, Rižana (Fig. 6), Malinska and Mirna rivers in the northern Istra Peninsula in particular (see Bogutskaya and Zupančič 2010) by having a shallower body, the body depth at the dorsal-fi n origin being 21-25% SL (means 22.5 and 23.4 in the Boljunšćica and Pazinčica samples, respectively) while in S. squalus the body is deeper, 23-28% SL (mean 25.4 in the Osapska sample, from Bogutskaya and Zupančič 2010). Th e head is longer in S. janae than in S. squalus: in S. janae the head length is 28-32% SL (means 28.2 and 29.1 in the Boljunšćica and Pazinčica samples, respectively) in contrast to 25-29% SL in S. squalus (mean 26.5 in the Osapska sample). Th e depth of the cau-   (Fig. 6). Additionally, the Boljunšćica and Pazinčica specimens have a straight oblique mouth cleft with a distinct angle at the lower jaw-quadrate junction, similar to the condition seen in S. janae and in contrast to S. squalus, which has a rounded snout and a shorter mouth cleft, curved in its anterior part and in a more horizontal position ( Fig. 6; Bogutskaya and Zupančič 2010: fi gs 4a, b, 9). Th e Boljunšćica and Pazinčica specimens are further distinguished from S. squalus by vertebral counts having 44 total vertebrae (Fig. 4), vertebral formulae being 24+20 or 25+19 vs. commonly 43 and 25+18, and a silvery colouration vs. darker colouration with brownish or slight bronze tones in S. squalus. Squalius zrmanjae (Zrmanja and Krka Rivers) is another species geographically close to S. janae. A morphological comparison between the two species and between S. janae and S. illiricus distributed in Cetina and Krka can be found in Bogutskaya and Zupančič (2010), and some distinguishing characters are summarised in the key below.

Distribution of chubs in Istra Peninsula
Th e known range of S. janae (Fig. 1) includes Dragonja River (type locality) draining to the west, Pazinčica River that used to fl ow westwards but at present terminates in a cave at Pazin, and Boljunšćica River that used to fl ow southwards but now ends in canals in an area that was formerly Čepić Lake. We do not have materials from Raša River that is adjacent to the Boljunšćica and is now connected by canals to the latter. Th ere are no historical samples known to us to decide upon the identity of the native Raša chub.
In rivers of north Istra (Osapska Reka, Rižana, Malinska, and Mirna) occurs a chub, which was identifi ed by Bogutskaya and Zupančič (2010) as S. squalus, a species distributed widely further north-and westwards in the Adriatic Basin. Th us, in the north-west of the Istra Peninsula the range of S. janae is interrupted by Mirna River, which is now inhabited by S. squalus. Th is may be explained as a historic indication of a direct connection of the Mirna to the Palaeo-Po system but a relatively recent introduction cannot be excluded either. To the east, Rječina River is the stream closest to Istra Peninsula, fl owing into the Adriatic Sea at Rijeka. No chubs (Squalius) are known from this river (Šprem 2006). Further southwards, there is the Dubračina, a small, now endorheic river system, with the Tribalj Reservoir, which is inhabited by a probably introduced Alburnus, a Rutilus, and introduced Cyprinus carpio. Th ere is a chub in Dubračina (Zupančič's data) though no specimens were collected to check their identity. Th ere are no published data indicating that a chub occurs in Ričina, a river located further down and fl owing into the Adriatic at Novi Vinodolski. Th ere are no native Squalius species in the entire Lika region endorheic drainages (Lika, Jadova, Otuča and some others) that lie south and west from Velika Kapela and Mala Kapela mountains southwards to Gračac. Squalius cephalus was introduced from the Danube to the Lika River (specimens in PZC). Further southwards, the only species of Squalius, S. zrmanjae, inhabits Zrmanja River, and this species together with S. illyricus and Squalius sp. occurs in the Krka River drainage.
Th e data presented above on the distribution of Squalius chubs in the northern Adriatic Basin support the assumption by Bogutskaya and Zupančič (2010) that the range of S. janae encompassing most of Istra Peninsula, except for its north-western section, is determined by the geology of two fl ysch basins forming the base of the Istra Peninsula. Th ese basins are the Trieste Flysch Basin and the Pazin Flysch Basin bordered by limestone areas of the Buje and West Istria anticlines in the west and the Čičarja and Učka mountain ranges, which belong to the Dinaric Alps, in the east (Fišer et al. 2006;Babić et al. 2007). It is well known and discussed in the literature (e.g. Lindberg 1972;Bianco and Miller 1990;Holčík and Mrakovčić 1997) that during several marine regressions including the most recent ones, for example during the last Würm glacial maximum, rivers of both slopes of the Adriatic basin used to form an extensive drainage of the Palaeo-Po-Isonzo (e.g. CLIMAP 1976, Rodić 1981. However, the distribution pattern of diff erent fi sh species, the occurrence of a certain number of endemic species in particular, does not support a simple model of fi sh dispersal within the Palaeo-Po's tributaries. Indeed, the distributional pattern of most fi sh taxa stands against a hypothesis that assumes a complete Palaeo-Po habitat connectivity and indicates a disrupted surface palaeohydrography that was heavily fragmented by karstifi cation in the whole Dinaric area. A similar assumption has been made for a number of taxa other than fi sh inhabiting the Dinaric Karst (Sket 2002, Trontelj et al. 2007).

-
No silvery tint in life colouration, overall colouration with brownish or bronze tones; scales not easily lost; iris, pectoral, pelvic and anal fi n pigmentation with reddish or orange shades; head length 26-29% SL, lower jaw length equal to or shorter than caudal peduncle depth; slightly curved mouth cleft .