Mugil trichodon Poey, 1875

Menezes, Naércio A., Nirchio, Mauro, Oliveira, Cláudio De & Siccharamirez, Raquel, 2015, Taxonomic review of the species of Mugil (Teleostei: Perciformes: Mugilidae) from the Atlantic South Caribbean and South America, with integration of morphological, cytogenetic and molecular data, Zootaxa 3918 (1), pp. 1-38 : 31-34

Mugil trichodon Poey, 1875

( Fig. 12 )

Mugil trichodon Poey, 1875: 66 (type locality: Cuba, no types known); Cervigón, 1992: 878 (northern coast of South America); 1993: 384 ( Venezuela; diagnosis, biological and ecological data); Thomson, 1997: 494 (only specimens originating from southern Caribbean area; description); González Bencomo et al., 1997: 159 ( Venezuela); Harrison, 2003: 1085 (only specimens from southern Caribbean; habitat, biology, and fisheries; distribution); Harrison et al., 2007: 96–97 (only specimens originating from Atlantic southern Caribbean; as comparative material examined)

Diagnosis. Mugil trichodon differs from congeners from the study area except M. curvidens and M. liza in having the anal fin with 3 spines and 8 branched rays in adults or 2 spines, 1 unbranched ray and 8 branched rays in juveniles (vs. 3 spines and 9–10 branched rays in adults or 2 spines, 1 unbranched ray and 9–10 branched rays in juveniles). Mugil trichodon is distinguished from M. curvidens by the presence of slightly curved teeth on upper jaw, 1 unbranched ray and 7 branched rays in the second (soft) dorsal fin in adults or 2 unbranched and 6 branched rays in juveniles (vs. teeth on upper jaw strongly cruved, 1 unbranched ray and 8 branched rays in adults or 2 unbranched rays and 7 branched rays in juveniles) and from M. liza in having the second dorsal and the anal fins scaled except distally (vs. second dorsal and anal fins with the scales restricted to an anterior basal portion and a single, sometimes incomplete posterior basal portion).

Description. Morphometric data presented in Table 13 . Maximum examined body length 213 mm SL. Body elongate, compressed and moderately deep compared to congeners. Greatest body depth at vertical through spinous dorsal-fin origin. Dorsal profile of head and body convex along tip of snout, slightly convex from vertical through anterior border of orbit to spinous dorsal-fin origin, straight from this point to soft dorsal-fin origin, slightly convex from this point to caudal peduncle, slightly concave along caudal peduncle. Ventral profile of head and body convex from tip of lower jaw to termination of base of anal fin, straight to slightly concave from this point to vertical through caudal-fin base. Orbital diameter much larger than snout length. Eye covered with adipose tissue, except for narrow oval-shaped central area in adult specimens. Adipose tissue almost absent in specimens smaller than 30–35 mm SL.

Anterior spinous dorsal fin with 4 slender spines connected by membrane, n = 17. Spinous dorsal fin-origin about midway between snout tip and caudal-fin base. Posterior soft dorsal-fin rays ii,6, in specimens smaller than 30–35 mm SL, i, 7 in adults, n = 17. Unbranched pectoral-fin rays ii, first ray much smaller than second; branched rays 15, n = 17. Tip of pectoral fin not reaching vertical through spinous dorsal-fin origin, extending to vertical through about middle of pelvic fin. Pelvic fin rays with I,5. Tip of pelvic-fin reaching vertical through spinous dorsal-fin origin. Anal fin rays with II,i, 8 in specimens smaller than 30–35 mm SL, III, 8 in adults, n = 17.

Mouth subterminal. Tip of maxilla extending slightly beyond vertical through anterior border of orbit. Teeth unicuspid, spatulate with slightly curved tips ( Fig. 12 A), visible with naked eye in large specimens. External teeth on upper lip larger that scattered fewer inner teeth. Single row of unicuspid close set finer teeth smaller that upper lip teeth.

Scales spinoid, spines rudimentary on surface of scales and not projecting posteriorly on scale margin ( Fig. 12 B). Transverse scale rows from dorsal limit of pectoral-fin base to caudal-fin base 30–33, 30.7, n = 17. Horizontal scale rows from spinous dorsal-fin origin to pelvic-fin origin 11–12, 11.05, n =17. Horizontal scale rows around caudal peduncle 15–16, 15.2, n = 17. Soft dorsal and anal fins densely scaled except for narrow scaleless distal area. Basal portion of pectoral fin fully covered by small scales extending between interradial membranes but not reaching distal margin of fin. Modified axilla scale dorsal to pectoral-fin base 2.8 times as long as pectoral fin in 216 mm SL specimen (LBP 6066). Modified axilla scale dorsal to pelvic fin 2.5 times as long as pelvic fin in same specimen.

Color in alcohol. Body dark dorsally, dark color fading ventrally towards midlateral region of body, whitish on abdominal region. Pelvic and anal fins pale with few scattered dark chromatophores. Spinous dorsal, soft dorsal, pectoral and caudal fins profusely covered with dark chromatophores. Distal margin of caudal fin darker than remaining parts of fin. Anterodorsal basal portion of pectoral fin with relatively large, dark spot extending over basal portions of unbranched rays and 10 dorsalmost branched rays.

Cytological and molecular data. Mugil trichodon has a 2n=48 karyotype with entirely acrocentric chromosomes (arm number, NF+48). The chromosomes gradually decrease in size such that the homologues could not be clearly distinguished, except pair 15, which shows a conspicuous secondary constriction. Heterochromatin blocks are mostly restricted to the centromeric regions of all chromosomes and some are more obvious than others. The chromosome pair characterized by a secondary constriction shows short heterochromatic arms. NOR signals correspond to secondary constriction and show significant intraindividual size variations between both homologous chromosomes ( Nirchio et al., 2005).

Molecular analyses showed that the number of nucleotide differences between M. trichodon and the remaining analyzed species range between 46.7 to 68.0 (16S) and 94.5 to 116.9 (COI) (tables 2 to 5). The genetic distance between this species and the remaining analyzed ones ranged from 0.092 to 0.128 (16S) and 0.172 to 0.210 (16S) and 0.172 to 0.210 (tables 6 and 7). The dendrogram in Figure 1 indicates that this species is genetically most similar to M. rubrioculus , Mugil sp. and M. brevirostris .

Distribution. Within the study area Mugil trichodon occurs only in the southern Caribbean area ( Fig. 4 ) and along the Venezuelan coast was found together with M. margaritae , M. incilis and M. rubrioculus . Previous records of the species from Brazil ( Menezes, 1983:3, Menezes et al., 2003: 65) are based on specimens actually belonging to Mugil curvidens Valenciennes.

Discussion. The results obtained herein indicate that one of the problems involving mugilid taxonomy is related to previous identification of widely distributed species based on the assumption that morphological characters are very similar across the family ( Thomson, 1997). However, given that most, if not all, the species live in coastal waters and are strongly dependent on local low salinity for growth and gonad maturation before moving off-shore to spawn and returning to coastal lagoons and other estuarine water bodies ( Marin et al., 2003; González- Castro et al., 2011), it seemed unlikely that they would undertake the extensive migrations and have intense gene flow among populations from distant areas, which is prerequisite for the existence of single species in different portions of the world oceans. The best example is Mugil brevirostris (Ribeiro) recognized in the present paper as a species found along the Brazilian coast instead of Mugil hospes Jordan & Culver (not Jordan and Cuvier as it appears in Harrison, 2003), a species originally described from the Pacific Ocean (Sinaloa, Mexico) and that was previously applied to this population ( Thomson, 1997; Harrison, 2003), simply because the differences between the two taxa were neglected.

The almost worldwide distribution of Mugil cephalus Linnaeus and the wide distribution in Mugil curema Valenciennes indicated by Thomson (1997) has been recently questioned in molecular studies based on the analysis of mitochondrial-DNA sequences ( Durand et al., 2012). Morphological ( Menezes et al., 2010) and molecularbased studies ( Siccha-Ramirez et al., 2014) demonstrated that Mugil liza Valenciennes is the name that must be used for the lebranche grey mullet in the Atlantic South Caribbean and South America, and so specific names such as M. cephalus and M. platanus do not apply to taxa found at that region. Mugil curema probably has a much more restricted distribution than that previously recognized ( Eschmeyer, 2014), and since the lectotype is a specimen from Bahia, Brazil (Harrison, 1993) the name should be used only for the species occurring in Atlantic South America until a more comprehensive revision is completed. Including specimens identified as Mugil curema from Argentina, based on morphometric characters, González-Castro et al. (2012) found differences between M. curema populations from Mexico and Argentina and suggested that the South America form identified as M. curema probably represents a new species.

Mugil incilis originally described from Guyana ( Hancock, 1830) and later reported from freshwater (Rio Xingu, northern Brazil) as Mugil xinguensis by Steindachner (1907) is instead restricted to the large estuarine area of the mouth of the Rio Amazonas extending north to Venezuela and south to Maranhão, Brazil, where the influence of low salinity waters still prevails. The reference to this species as occurring from South Caribbean to Argentina ( Siccha-Ramirez et al., 2014) is incorrect. This species has never been collected to date south of Maranhão and all the records south of this locality are probably due to Ribeiro’s (1915) reference to specimens from Sapucaia, Rio Parahyba (= Rio Paraíba), Rio de Janeiro on which he based his short description of the species. One of us (NAM) examined these and other specimens identified as M. incilis from Rio de Janeiro belonging in the MNRJ collection and found they were misidentified (see Remarks under Mugil incilis above).

Although there are no details about the karyotypes of two of the eight species studied herein ( Mugil brevirostris and M. curvidens ) the available cytogenetic data shows that diploid number, constitutive heterochromatin distribution and the striking differences in the location of the NORs allow to distinguish the other six species of the Atlantic South Caribbean and South America ( M. curema , M. margaritae , M. incilis , M. trichodon , M. liza and M. rubrioculus ). Two species are easily distinguished by their unique chromosome numbers ( M. curema with 2n = 28 and M. margaritae with 2n = 24) and although all the remaining caryotyped species have 2n = 48 chromosomes their karyotypes with distinct NOR-bearing chromosomes indicates a clear distinction among them showing that cytogenetic data are good markers for species of Mugil in the studied area.

The molecular studies on the other hand, involving the sequencing of the genes 16S (with slow evolutionary rates) and COI (with fast evolutionary rates) show that all the species from the Atlantic South Caribbean and South America can be readily distinguished from each other by the presence of a large number of fixed mutations and by high genetic distances. The present data was also compared with those from four samples of Mugil cephalus sequenced by Durand et al. (2012), taking into consideration that some authors suggested the occurrence of this species in the study area. We found that all the species sequenced herein can be distinguished based on molecular data. Lower differences were found when M. cephalus was compared with M. liza (Tables 2 to 7), the lowest numbers for different nucleotides being 4 (16S) and 15.3 (COI) and the lowest genetic distances 0.007 (16S) and 0.025 (COI). However, the lowest value of the genetic distance (2.5%) in the COI gene is slightly higher than the mean species level divergence (2%) found among fish species (Ward, 2009).

Figures

FIGURE 1. Mugil brevirostris, MZUSP 31446. A—Detail of mouth showing part of dentition. B—Detail of scales located on dorsal region of body below dorsal-fin base.

FIGURE 4. Map of central and northern South America showing collecting sites and type localities (white star and light blue star) of species of Mugil. Pink dot: Mugil trichodon, M. rubrioculus and M. incilis (Venezuela); white star: holotype of M. margaritae; white dot: M. margaritae; yellow dots: M. incilis (Brazil); green dots: M. rubrioculus (Brazil); light blue dot: M. brevirostris; blue star: neotype of M. brevirostris; red dots: M. curvidens. Some symbols may represent more than one locality.

FIGURE 12. Mugil trichodon, LBP 6066. A—Detail of mouth showing dentition. B—Detail of scales located on dorsal region of body below dorsal-fin base.

Tables

TABLE 13. Morphometrics of Mugil trichodon. Standard length in mm. Measurements through head length are percentages of standard length; last eight entries are percentages of head length.

Characters	n	range	mean	SD
Standard length	17	158.0–213.0	186.8
Body depth	16	023.8–029.1	0 26.6	1.3
Snout to dorsal-fin origin	16	073.2–076.8	0 75.0	0.9
Snout to pectoral-fin origin	17	026.0–028.7	0.27.3	0.8
Snout to pelvic-fin origin	17	038.2–041.2	0 40.1	0.7
Snout to anal-fin origin	17	069.6–073.2	0 71.0	1.1
Caudal peduncle depth	17	010.6–012.4	0 11.4	0.4
Caudal peduncle length	17	017.5–021.0	0 19.7	1.1
Pectoral-fin length	13	016.2–021.0	0 19,4	1.3
Pelvic-fin length	15	013.6–017.0	0 15.4	1.0
Head length	16	025.1–027.4	0 26.0	0.7
Head width	16	065.8–074.3	0 69.1	2.1
Head depth	17	078.2–086.8	0 82.4	2.7
Lip thickness	16	006.0–010.0	0 0 7.6	1.0
Mouth width	17	024.0–032.5	0 28.7	2.0
Mouth depth	17	027.3–032.8	0 29.4	1.7
Horizontal orbital diameter	16	025.8–032.4	0 29.2	1.9
Snout length	17	019.6–025.8	0 24.0	2.0
Upper jaw length	16	039.5–044.0	0 41.8	1.3