Rocio octofasciata ( Regan, 1903 ) Juan J. Schmitter-Soto, 2007

Juan J. Schmitter-Soto, 2007, A systematic revision of the genus Archocentrus (Perciformes: Cichlidae), with the description of two new genera and six new species., Zootaxa 1603, pp. 1-78 : 57-59

Rocio octofasciata ( Regan, 1903) , n. comb.

Figures 22-24

Heros cyanoguttatus (part. et non Baird & Girard), Evermann & Goldsborough 1902: 157 (misidentification).

Heros octofasciatus Regan, 1903: 417 (original description).

Cichlasoma octofasciatum , Meek 1904: 218 ( new combination).

Cichlasoma hedricki Meek, 1904: 208 (junior synonym).

Cichlosoma biocellatum Regan, 1909: 234 (junior synonym).

Parapetenia octofasciata , Allgayer 1989: 26 ( new combination).

‘Cichlasoma’ octofasciatum , Kullander 1996: 151 (incertae sedis).

Archocentrus octofasciatus , Schmitter-Soto 1998: 154 ( new combination).

Nandopsis octofasciata , Burgess2000: 48 ( new combination [as octofasciatum ]).

Holotype. MHNG 665.55 , 39 mm SL (Fig. 23), A. C. Buller, Feb. 1, 1866. “ Rio de Sarabia” (Cosamaloapan, according to the original label), Coatzacoalcos drainage , Veracruz, Mexico. No paratypes, but originally mixed, unlabeled, in same jar with eight more specimens (see Remarks).

Diagnosis. No unique autapomorphies. Spots on sides smaller than scales, aligned in ca. 15 regular series (vs. not clearly aligned); abdomen predominantly whitish or greyish in life (also in R. gemmata , vs. reddish in R. ocotal ); ventral angle of articular, acute (vs. right); first neural spine oriented rostrad (vs. caudad); circumpeduncular scales as few as 17 (vs. always more than 19); distance from caudal esophageal loop in gut to esophagus always greater than 24% gut length (vs. less than 16%).

Description. D. XVII-XIX,8-10, modally 10 (7 of 174 with 11 dorsal fin rays); A. VIII-IX,7-9, modally 8 (2 of 177 with 6 or 7 anal-fin spines); pectoral 14-16. Gill rakers rounded, distally expanded or trapezoidal, sometimes bifid and even trifid, sometimes serrated. Scale rows on cheek modally 6 (4-7, 12 of 68 specimens with 7 scale rows); predorsal scales modally 12; pored lateral-line scales (not counting scales overlapping between the two segments of the lateral line) 27-30; scales from lateral line to origin of dorsal fin 4-5.5; scales from lateral line to base of first dorsal-fin ray 3-4 (further meristic data appear in Table 3).

Largest specimen examined, 217 mm SL, but reported to 250 mm SL ( Page & Burr 1991). Body usually slender, deeper in young (39-51% of SL). Head length 33-42% of SL; orbital diameter usually 21-25% of head length (up to 31% of head length in juveniles -further morphometric data appear in Table 4). Head profile convex, straight, concave on orbits. Maxilla reaching only a horizontal line, not a vertical, from orbit. Corners of lower lip slightly tapering, if at all.

Pectoral and pelvic fins nearly always reaching caudad beyond 1st or 2nd anal-fin spine (not reaching in 2 of 127 specimens). Filamentous rays of dorsal fin to mid-caudal fin. Scales between dorsal and anal fin rays in one row, up to 6 scales long.

Gut simple, anal and anterior esophageal loops adjacent; gut length shorter or subequal to standard length of fish. Genital papilla thick, cylindrical, notched or triangular-tipped, sometimes sunk, longer than broad, medially swollen, somewhat crenulate at tip, sometimes rather vase-shaped; pigmented just in basal half and margins, if at all.

Stripe from snout to eye diffuse or absent; suborbital streak narrow, with pointed ends (not always visible in preserved specimens). Eye color varied: yellowish, greyish, bluish, coppery, reddish. Bars on sides sometimes diffuse, sporadically dorsally or medially coalescent. Lateral blotch oval to squarish, joining 3rd and 4th bars, sometimes not discernible from longitudinal stripe. About 15 rows of light spots on sides, centered in each scale, not always visible; breast region olive. Axil of pectoral fin with same coloration as breast or dusky; base of pectoral fin paler than surrounding region. For life colors, see Fig. 24.

Distribution. From Río Ulúa, Honduras ( Lee et al. 1980), to tributaries of Río Actopan, Veracruz, Mexico ( Greenfield & Thomerson 1997) (Fig. 22). Introduced in several localities in northern Veracruz ( Obregón et al. 1994).

Remarks. Contra Regan’s (1903) original designation (and S. Fisch - Müller ’s opinion, pers. com.), Mahnert(1976: 44) treated the nine specimens originally in MHNG 665.55 as syntypes. However, there is little doubt that the largest individual is the holotype. The confusion was likely created by Regan (1904, 1905) himself, when he added material to his successive redescriptions, somewhat more accessible than the original description.

Cichlosoma biocellatum , based on an aquarium specimen, is a synonym, its type locality (“ Río Negro at Mañaos [sic], Brazil”) certainly in error.

Cichlasoma hedricki was synonymized by Regan (1905), barely one year after Meek’s (1904) description, without explicit reasons. The action was followed by most authors. Topotypes of C. hedricki (from the upper Papaloapan) tend to be deeper-bodied than other R. octofasciata , and the presence of a prominent lingual cusp in their symphysial teeth (fig. 15a in Schmitter-Soto, in press) tends to be more constant, but the forms merge imperceptibly towards the type locality of R. octofasciata in the Coatzacoalcos, the drainage immediately to the south of Papaloapan.

In print, Meek (1904) called the type locality of C. hedricki “Obispo, Veracruz;” however, his manuscript label in the paratype jar at UMMZ (176671) reads “ Rio Obispo , Oaxaca.” The eight paratypes in this lot (labelled as such by Meek himself) should probably be added to the 44 specimens mentioned by Eschmeyer (2005); however, on the other hand, the original description mentions only one “type,” thus rendering the literature (and paratype holdings at UMMZ and FMNH) in error.