MORPHOLOGICAL DEFORMITIES AND ATYPICAL COLOUR PATTERN IN THORNBACK RAY , RAJA CLAVATA ( ELASMOBRANCHII : RAJIFORMES : RAJIDAE ) , FROM IZMIR ( TURKEY , AEGEAN SEA , EASTERN MEDITERRANEAN )

Two specimens of the thornback ray, Raja clavata Linnaeus, 1758, from Turkish waters, exhibiting deformities of disc and tail are described in the present paper. One of these specimens had a colour pattern similar to type 7, vermiculated, it could constitute a transitional step between Raja clavata and its close relative species Raja maderensis Lowe, 1838.


INTRODUCTION
Thornback ray, Raja clavata Linnaeus, 1758, is known to occur along the eastern Atlantic coast, from Scandinavia, around the British Isles, France, and Portugal to the Strait of Gibraltar (Quéro et al. 2003).Further south, the species R. clavata is recorded off Morocco (Lloris and Rucabado 1998), Mauritania (Maurin and Bonnet 1970), and off the South African coast (Smith and Heemstra 1986).The thornback ray is also reported throughout the Mediterranean including the eastern areas (Golani et al. 2006), and it occurred in the Black Sea (Bilecenoğlu et al. 2014) Raja clavata is distributed but not targeted in Turkish seas, and rather discarded at sea by fishermen.The species has low value and is not consumed locally, but it is sometimes exported to foreign countries such as Greece and United Kingdom.The total catch reached 116 t in 2016, and the most abundant landings were reported from the Sea of Marmara (63%), Aegean Sea (28%), and Black Sea (9%) (Anonymous 2017).Captures of thornback rays remained to be very rare in Izmir Bay (Aegean Sea), and during experimental trawl surveys carried out between 2006 and 2009, the total catch of R. clavata was only 3 kg from 116 hauls (Gurbet et al. 2013).
Investigations were conducted during a decade in Turkish waters focusing on elasmobranch species and constituted a larger research project.Some of the results of the project have already been published (Akyol and Capapé 2014, Akyol et al. 2017a, 2017b).Presently reported two R. clavata specimens, exhibiting tail abnormalities, were also a part of the above-mentioned project.The aim of this paper was to give a description of each specimen following Capapé et al. (2015aCapapé et al. ( , 2015b) ) and comment on the origin of these atypical characteristics, and their role in the species growth.

MATERIAL AND METHODS
The first abnormal specimen of Raja clavata (Fig. 1) was caught on 30 September 2016, and a second (Fig. 2) on 7 February 2017; both at depths between 70 and 80 m by bottom trawl (44 mm mesh size in the cod end), off Cesme, Izmir (38°25′N and 26°16′E).During the study, all absolute measurements were recorded to the nearest millimetre while all weights to the nearest gram.The absolute measurements were also expressed in relative values-as the percentage of the disc width (DW).The measurements and the meristic counts were performed following the methodology of Quignard (1965) and Stehmann (1970), except for trunchal vertebrae and pectoral rays counted based on X-ray photographs.The count of nictitating lamellae followed Capapé and Quignard (1981) and Mnasri et al. (2009aMnasri et al. ( , 2009b)).Morphometric measurements and meristic counts were summarized in Table 1, including also a normal specimen for comparison with the abnormal ones.Additionally, the relation between the disc width (DW) and the total body weight (TBW), was calculated including 39 normal specimens and both abnormal specimens, all collected in the same area.This relation is used to assess if the abnormal specimens are able to develop in the wild (Froese et al. (2011).Disc width was chosen as a reference character rather than total length, as recommended by Clark (1926) for studies concerning batoid species.This relation was expressed in decimal logarithmic scales and correlations were assessed by least-squares regression, to obtain a linear regression.
All specimens were delivered to the laboratory for subsequent study.Identification to the species level was performed based on Capapé and Desoutter (1981) and Stehmann and Bürkel (1984).The two abnormal specimens and the normal specimen (see Table 1 Capapé et al. 2007).The general shape of the disc showed an asymmetry between the left and the right side (Fig. 1A).The former was more sinuous and pointed in its distal end while than the latter was rather rounded, especially in its distal end.The snout was conical and rounded, shorter than those observed in normal thornback rays (Fig. 1B).Both margins of the rostrum clearly showed a deformity, having the shape of a notch (Fig. 1B).It appears difficult to state about the origin of these deformities.The specimen was not entangled during its capture and therefore its deformity did not represent a healed scar, such as another specimen of R. clavata caught earlier in the area (Akyol and Aydin 2018).The first specimen escaped from gill nets or was discarded at sea after capture by fishermen because of the lack of its commercial value locally.A genetic origin of the abnormality cannot be totally ruled out, however, due to the fact that each notch was symmetrically arranged in each margin of the rostrum.The tail did not extend the length of the disc, but slightly deviated to the right side and strongly convoluted in the distal region, forming a large S shape, and X-ray photograph showed that in this region the vertebrae fused and could not be individually distinguished (Fig. 1C).Conversely, both dorsal and caudal fins were present and complete.The origin of the tail deformity could be accidental, probably occurring during the early period of the specimen life in the wild, or an abnormal development resulting from a genetic reason.
Second abnormal specimen of Raja clavata (Ref. ESFM-PIS/2017-03).The specimen measured 520 mm DW, 730 mm TL, and weighed 3411 g.It was also an adult female (Fig. 2A).The general shape of the specimen did not show an abnormality, however, the snout was minute and reduced (Fig. 2B), and the caudal fin was lacking at the distal tip of the tail, no scar was visible and this character could be considered as an abnormality (Fig. 2C).
Additionally, colour patterns of the dorsal surface were characteristic of type 7, vermiculated sensu Mnasri et al. (2009a), described as the colour of dorsal surface green covered with white spots, circles, half-moons or lunules forming vermiculations which extend to the edges of the disc.Such colour patterns are very similar to those from Madeira ray, Raja maderensis Lowe, 1838, according to Stehmann and Bürkel (1984).

DISCUSSION
Abnormalities of pectoral fins occur most frequently in batoid species (Ribeiro-Prado et al. 2008) and can reach up to 50% of cases (Orlov 2011).Such phenomenon is due to the fact that pectoral fins failed to fuse together in front of the head in the early development (Mnasri et al. 2010).The first specimen (Ref.ESFM-PIS/2016-14) exhibited a light asymmetry between left and right wing and evident notches on each site of the rostrum, constituting the first case known to date in skates.Bensam (1965) noted that embryonic deformities could be caused by intrauterine pressure exerted by other embryos.Rajid species are oviparous and the embryonic development occurs in egg case laid by females, consequently the origin of pre-natal abnormalities would be related to mutation or other developmental irregularities, and the role of unfavourable conditions such as large exposure to pollutants cannot be totally ruled out (Ribeiro-Prado et al. 2008).The study area, Izmir Bay and its vicinity, has large harbours, ship dismantling areas (i.e., Aliaga coast), and Gediz River delta, therefore, a significant enrichment in trace metals in sediments is regularly reported (Kucuksezgin et al. 2006, Kucuksezgin 2009, Pazi et al. 2017).Additionally, trace metals can decrease the collagen synthesis, cause protoplasmic poisoning, and change the integrity of bones (Iguchi andSano 1982, Jawad et al. 2017).
Loss of part or total tail due to injury is quite frequent especially in stingrays, but rather rare in skates (Templeman 1965, Mnasri et al. 2009b, Orlov 2011).However, it was difficult to state if such injury occurred recently or during the earlier life stages of the specimen.Conversely, the distal end of the second specimen (Ref.ESFM-PIS/2017-03), described in the present paper, did not exhibit a scar, so the lack of anal fins, was not the consequence of an injury and rather suggested a teratological case.Similar patterns could be taken into consideration about the tail abnormality observed in the first specimen (Ref.ESFM-PIS/2016-14).
Abnormalities of tail in skates species are rather rare, less than 4% according to Orlov (2011), and few instances were recently reported (Mnasri et al. 2010, Capapé et al. 2015a, 2015b) In total accordance with Orlov (2011), the three specimens described herein were not considerably affected by loss and abnormalities of tail since did not affect their development and swimming activities, although percentages of tail lengths versus disc width are lower in abnormal specimens than in normal specimen (Table 1).It is well known that skates and rays generally use pectoral fins for locomotion.These observations are corroborated by the relation between the disc width and the total body weight, plotted in Fig. 3, which showed a positive allometry with b = 3.32, indicating that abnormal specimens reached similar biological parameters that the normal ones.However, the risk of mortality was probably greater in such abnormal specimens than those presenting a tissue loss or abnormalities of tails and could explain why the former were relatively less observed than the latter.
The colour patterns of the second specimen (Ref.ESFM-PIS/2017-03), in our opinion, constitutes a new case of type 7, vermiculated sensu Mnasri et al. (2009a) Ball et al. (2016) noted that genetic identification of specimens of R. maderensis and R. clavata was not sufficient to state about the fact that they are close but distinct species.The presently reported second specimen is closer to R. maderensis rather than R. clavata confirming the opinion of Ball et al. (2016) who noted that morphological differences between the two species are not in total accordance with the genetic distinctiveness.Following the opinion of Ball et al. (2016), type 7, vermiculated sensu Mnasri et al. (2009a), probably could be considered as a transitional "taxon" between Raja clavata and Raja maderensis.
Fig. 1. A. Raja clavata (Ref.ESFM-PIS/2016-14), general morphology, scale bar = 50 mm; B. Anterior region of the disc, with arrow 1 indicating snout and arrows 2 indicating notch on each side of the rostrum, scale bar = 50 mm; C. Tail with arrow indicating the abnormality, S shape, scale bar = 50 mm

Fig. 3 .
Fig. 3. Relation between the total body mass (TBM) and the disc width (DW) expressed in logarithmic scale for normal and abnormal specimens of Raja clavata collected from Turkish waters

Table 1
Absolute and relative morphometric measurements and meristic counts of two abnormal specimens and one normal specimen of Raja clavata collected from Turkish waters ABS = absolute measurement, REL = relative value of measurement expresses as percentage of disc width (DW).
, and thus increases the difficulty to separate Raja clavata