Ptychochromoides itasy, Sparks, John S., 2004

Sparks, John S., 2004, A new and presumably extinct species of Ptychochromoides (Teleostei: Perciformes: Cichlidae) from central Madagascar, Zootaxa 524, pp. 1-15 : 4-12

publication ID 10.5281/zenodo.157248

publication LSID


persistent identifier

taxon LSID

treatment provided by


scientific name

Ptychochromoides itasy

sp. nov.

Ptychochromoides itasy View in CoL , new species

( Figures 2–3 View FIGURE 2 View FIGURE 3 and Table 1 View TABLE 1 )

Ptychochromis betsileanus: Pellegrin, 1933: 144 View in CoL (in part; Lake Itasy View in CoL ); Arnoult, 1959: 72 (in part; Lake Itasy View in CoL ); Kiener, 1963: 166 (in part; region of Lake Itasy View in CoL ).

Ptychochromoides betsileanus: Kiener and Maugé, 1966: 77 View in CoL –80 (in part; description); Reinthal and Stiassny, 1997: 358 –362 (in part; redescription); Sparks and Reinthal, 2001: 120, 131 (in part; comparative material); Sparks, 2003: 327 (in part; comparative material).

Holotype: UMMZ 243393, immature female, 123.2 mm SL; Madagascar.

Paratypes: AMNH 233643, 1 specimen, immature male, 106.3 mm SL; Madagascar. MNHN 1919­11, 1 C&S specimen, 77.5 mm SL; Madagascar: central highlands: Region of Antananarivo: Lake Itasy . UMMZ 199409, 1 specimen, immature female, 91.2 mm SL; Madagascar.

Diagnosis. Ptychochromoides itasy is distinguished from congeners by the presence of a pronounced occipital hump, regardless of size, sex, or sexual maturity [vs. absence ( P. vondrozo ) or presence only in adult males ( P. betsileanus )], and also by a shorter predorsal length (39.3–40.2% vs. 40.4–46.2% and 43.7–49.3% SL in P. betsileanus and P. vondrozo , respectively), more elongate terminal dorsal­fin spine (14.1–16.4% vs. 11.4–13.9% and 11.1–14.0% SL in P. betsileanus and P. vondrozo , respectively), and shorter snout (33.2– 35.7% vs. 36.1–45.1% and 36.6–41.5% HL in P. betsileanus and P. vondrozo , respectively).

Scales in lateral line 6 34 (2), 35 (2), 36 (1), 37 (1) 2 35

Scales: lateral line to dorsal fin 8 5 (4), 6 (4) 3 5 (1), 5.5 (2) (H)

Dorsal fin 9 XIV 11 (3), XIV 12 (4), XIV 13 4 XIV 11 (1), XIV 12 (2), XIV 13 (1) (H)

(1), XV 12 (1)

Anal fin 9 III 9 (3), III 10 (6) 4 III 10 (3), III 11 (1) (H)

Vertebrae (pre­caudal+caudal) 7 14 + 14 = 28 (1), 15 + 14 = 29 4 15 + 14 = 29 (3), 15 + 15 = 30 (1) (H)

(5), 15 + 15 = 30 (1)

Gill rakers (lower limb) 10 10 (4), 11 (5), 13 (1) 3 10 (1), 11 (2) (H)

Description. Descriptive morphometric and meristic features for the new species and P. betsileanus are summarized in Table 1 View TABLE 1 . A relatively small, deep bodied, and laterally compressed Ptychochromoides . Jaws isognathous to slightly retrognathous. Dorsal body profile more or less straight posterior of occipital hump. Ventral body profile moderately convex. Predorsal profile strongly convex, even in smallest specimen examined (UMMZ 199409, 91.2 mm SL), an immature female, owing to prominent adipose occipital hump in both sexes. Radiographs reveal an accumulation of adipose tissue dorsal to supraoccipital and frontal bones ( Sparks & Reinthal 2001: fig. 5b). Snout relatively straight. Mouth small and oblique, lips not fleshy. Caudal peduncle short, laterally compressed, and deep. Supraoccipital crest short and deep. Dorso­medial gap present between frontal and supraoccipital bones; suture between bones pronounced ( Sparks & Reinthal 2001: fig. 5b).

Vertebrae. Total vertebral count 29 or 30 [formulae: 15 + 14 (3) or 15 + 15 (1)].

Teeth. Buccal dentition comprised of closely set, distally expanded, bilaterally symmetrical bicuspid teeth. Oral teeth range from weakly to strongly bicuspid. Dentition covers slightly more than 3/4 of premaxillary arcade, and approximately anterior 3/4 of dentary. Outer row teeth enlarged and graded in size laterally. Anteriorly, outer row teeth procumbently implanted in both upper and lower jaws. Both upper and lower jaws with two to four inner rows of smaller, bilaterally symmetrical, bicuspid teeth. Inner row teeth generally not expanded at crown. Inner row teeth usually symmetrically bicuspid, but may be conical and unicuspid.

Pharyngeal toothplates. Lower pharyngeal jaw (LPJ) robust, covered both anteriorly and postero­laterally with weakly hooked, bicuspid teeth. Posteriorly, teeth more strongly hooked and bicuspid. Teeth increasingly robust medially, becoming quite enlarged but not molariform, particularly postero­medially on LPJ. Postero­medial teeth enlarged and cylindrical with single, short median cusp at posterior margin (i.e., markedly enlarged hooked and bicuspid teeth). LPJ wider than long and fully sutured with weak interdigitations along postero­ventral margin. Teeth on pharyngobranchial 3 hooked and bicuspid laterally, robust and enlarged (but not molariform) medially. Single row of dentition present on elongate 2nd pharyngobranchial toothplate, characteristic of genus. Teeth on 2nd pharyngobranchial toothplate elongate, hooked and bicuspid. Well­developed toothplates present on dorsal surface of 4th ceratobranchial bones, confluent with outer row of rakers. Dentition on 4th ceratobranchial toothplates ranges from elongate, conical, and unicuspid (laterally) to weakly hooked and bicuspid (medially).

Gill rakers. Ten or 11 elongate and triangular gill rakers present along lower limb of first gill arch, including raker in angle of arch. Ceratobranchial rakers on first gill arch denticulate medially. Ceratobranchial rakers on gill arches 2 through 4 strongly denticulate dorsally. Five to seven elongate rakers present on first epibranchial bone. Low count of five corresponds to alcoholic specimens in which reduced dorsalmost rakers possibly occluded by tissue. Pseudobranch exposed and gill­like.

Squamation. Body covered with large and regularly imbricate scales from approximately mid­orbit (dorsal margin) to origin of caudal fin. Scales on chest and belly somewhat reduced in size and embedded. Body scaled except snout, lachrymal, and preopercle. Scales on head, cheek, nape, opercle, ventral chest, and belly cycloid. Cheek scales in two to three rows. Lateral body scales weakly ctenoid anteriorly, becoming increasingly ctenoid posteriorly. Scales on caudal fin, posterior of hypural flexure, cycloid and reduced in size. Cycloid scales of reduced size also present in one or two rows along dorsal­ and anal­fin bases. Lateral line canal and pores well developed.

Fins. Dorsal fin with XIV spines and 11–13 soft rays. Anal fin with III spines and 10 or 11 soft rays. First anal spine very short, whereas second and third spines elongate and practically equal in length. Soft dorsal and anal fins produced, trailing margins filamentous and extending well past caudal­fin origin. Origin of dorsal fin located approximately at vertical through pectoral­fin insertion. Origin of pelvic fins located well posterior of vertical through pectoral­fin insertion. Pelvic fins short, extend to or just past anus when adducted. Caudal fin moderately emarginate.

Additional remarks on osteology. Suture between frontal bones and supraoccipital with dorsal gap and expansion (= convexity) of frontal bones. Lachrymal of plesiomorphic condition within Cichlidae , being subdivided into two broadly contiguous plates (i.e., lachrymal and ‘primitive second infraorbital’ of Cichocki, 1976). Lachrymal with four marginally­directed canals. Bones composing infraorbital series robust and well ossified. Two arms of first epibranchial bone of relatively same length. Median frontal pores of neurocranium (NLF0 of Barel et. al. 1977) separate, not coalesced in the midline. Exoccipital foramina present, however, these excavations much smaller and less developed than in etropline cichlids ( Paretroplus + Etroplus ).

Coloration in life. Not known.

Coloration in preservative. Overall ground coloration golden brown and somewhat blotchy, with no distinctive markings ( Figs. 2–3 View FIGURE 2 View FIGURE 3 ). Notably, all available specimens appear relatively faded in preservative. Body slightly darker along dorsal midline. Fin rays grayish­black.

Habitat and distribution. Of the four specimens in the type series, only one is known for certain to have been collected in Lake Itasy (MNHN 1919­11, 77.5 mm SL, C&S). Based on morphological similarities to the unquestionable Lake Itasy specimen, the other three members of the type series (AMNH 233643, UMMZ 199409, UMMZ 243393) are presumed also to have been collected from Lake Itasy or the surrounding region. The three latter specimens were transferred from MNHN (formerly MNHN 1907­105, 106, 107) to UMMZ as gifts in the early 1970s and, other than 'Madagascar', are accompanied by no additional collection locality information. Likewise, no further collection locality information was available for these specimens in the MNHN records.

Based on available habitat data for all nominal species of Ptychochromoides , it appears that these cichlids are restricted to relatively undisturbed highland rivers, streams, and lakes that are well­oxygenated and that remain cool throughout the year ( Kiener 1959, 1963; Reinthal & Stiassny 1997; Sparks & Reinthal 2001). Ptychochromoides are absent from regions of the island where there has been a high degree of erosion and siltation as a result of deforestation and/or the introduction of numerous exotic species ( Kiener 1959; Sparks & Reinthal 2001). Members of the genus are also absent from lowland and coastal areas. Unlike some species of the closely related genus Ptychochromis , members of Ptychochromoides do not enter brackish water habitats.

Regrettably, it appears that the Lac Itasy population of Ptychochromoides (i.e., P. itasy ) is extinct. Despite claims of its persistence in the upper reaches of the Tsiribihina and Betsiboka drainages by local residents, repeated efforts over the past decades to collect additional material of P. i t a s y from either basin have been unsuccessful (P. Loiselle, pers. comm.). Fisheries researchers studied Lake Itasy for a number of years ( Kiener 1959, 1963; Kiener & Maugé 1966). According to Kiener (1959: 3, 1963: 167), Ptychochromoides were once so abundant in Lac Itasy that they accounted for a significant proportion of the overall catch from that basin. Concomitant with a decline in water quality due to human mediated disturbances, particularly overfishing and the introduction of a number of exotic species, was the rapid decline of Ptychochromoides and other endemic fish species throughout the region ( Reinthal & Stiassny 1991; de Rham & Nourissat 2002). De Rham and Nourissat (2002) hypothesize that Ptychochromoides was extirpated from the Lake Itasy (central highland) region by the early 1960s. As the small size of the type series attests, extremely little material from the region was preserved and deposited in museum collections, despite numerous claims regarding the former abundance of the species. This unfortunate situation underscores the importance of preserving and depositing voucher material for future study. Lac Itasy , a large crater lake, and the surrounding region are highly disturbed (de Rham & Nourissat 2002; M. Stiassny & Paul Loiselle, pers. comm.), and these basins are now home to a number of exotic species ( Reinthal & Stiassny 1991). No specimens of Ptychochromoides have been collected from the lake or surrounding basins in several decades, despite multiple attempts by various research groups ( Reinthal & Stiassny 1991, 1997; de Rham & Nourissat 2002).

Relationships and comparisons. Ptychochromoides itasy represents the fourth species of the genus to be described. All species of Ptychochromoides exhibit allopatric distributions ( Fig. 1 View FIGURE 1 ). Given that Ptychochromoides is not monophyletic, and that P. katria is recovered as the sister taxon to Ptychochromis in recent phylogenetic studies (Sparks 2003, 2004), generic comparisons are restricted to members of a clade comprising Ptychochromoides betsileanus , P. vondrozo , and P. i t a s y. Based on the analysis of nucleotide characters, Oxylapia polli is recovered as the sister taxon to this latter clade (Sparks 2003, 2004). Based on morphological comparisons, the new species is the hypothesized sister taxon to P. betsileanus , with which it shares the following features (that are absent in P. vondrozo ): 1) A deep supraoccipital crest in which there is a marked concavity between the frontal bones and supraoccipital in lateral view ( Sparks & Reinthal 2001: fig. 5). 2) The soft dorsal and anal fins are produced and the trailing margins are elongate and filamentous, extending well beyond origin of the caudal fin ( Figs. 2–4 View FIGURE 2 View FIGURE 3 View FIGURE 4 ). 3) Origin of the dorsal fin is located slightly posterior to a vertical through pectoral­fin insertion, whereas in P. vondrozo dorsal­fin origin is located well posterior to a vertical through pectoral­fin insertion. In general, P. betsileanus and P. i t a s y are also deeper bodied than P. vondrozo (45.2– 52.5% SL vs. 39.4–47.3% SL, respectively).

A sheared principal components analysis of 13 morphometric variables separated P. betsileanus , P. i t a s y, and two anomalous specimens (MNHN 1907­104, Ptychochromoides sp.) into three essentially non­overlapping clusters ( Fig. 5 View FIGURE 5 ). Other than Madagascar, no more precise collection locality information is available for these latter two specimens in the MNHN database. Sheared PC2 loaded heavily for preorbital depth and upper jaw length, whereas sheared PC3 loaded heavily for predorsal length and orbit diameter ( Table 2 View TABLE 2 ). Discrimination of P. betsileanus and P. i t a s y occurred along sheared PC 3, whereas sheared PC2 clearly separated P. betsileanus and P.i t a s y from the two anomalous specimens (MNHN 1907­104) ( Fig. 5 View FIGURE 5 ). Both sheared PC2 and sheared PC3 discriminate P. itasy from the two MNHN specimens. Only a single proportional measurement (head length: 35.1–35.2% SL), a variable that does not load heavily on either sheared PC, separates the MNHN specimens ( Ptychochromoides sp.) from P.betsileanus (30.0–33.8% SL) and P. itasy (31.1–32.5% SL). Thus, at the present time it seems premature to diagnose these two specimens (MNHN 1907­104) as new on the basis of this single proportional measurement.

In addition to the features listed in the differential diagnosis, P. i t a s y is further distinguished from P. vondrozo by a deeper caudal peduncle (15.8–17.5% vs. 12.9–14.8% SL), shorter head length (31.1–32.5% vs. 33.6–36.5% SL), longer pelvic fin (27.4–27.6% vs. 23.6–26.7% SL), and fewer scales in lateral series (35 vs. 36–40) ( Figs. 2–3 View FIGURE 2 View FIGURE 3 and 6 View FIGURE 6 ).

The type specimens of P. betsileanus described by Boulenger (1899) were supposedly collected from within the Betsileo region, ranging from southern central to southern Madagascar, however specific locality information is lacking. All catalogued specimens of Ptychochromoides in museum collections were examined during this study. Available collection locality data indicates that only a single immature specimen (MNHN 1919­11, 77.5 mm SL, C&S) can be determined with certainty to have originated from the Lac Itasy (central highland) region. This is surprising as Ptychochromoides were reported to have been abundant in Lake Itasy up until the late 1950s ( Kiener 1959, 1963; Kiener & Maugé 1966). Based on morphological similarities with this individual, it is concluded that the three additional specimens of the type series of P. i t a s y that lack specific collection locality data are also from the central highlands of Madagascar (i.e., Lake Itasy or the surrounding region).

Recently collected P. betsileanus from south­central Madagascar differ from P. i t a s y in possessing robust molariform dentition on both the upper and lower pharyngeal toothplates (UPJ and LPJ respectively), whereas in P. i t a s y, UPJ and LPJ dentition may be quite enlarged and robust, but never molariform. Interestingly, members of the type series described by Boulenger (1899), supposedly collected from the Betsileo region ranging from southern central to southern Madagascar, also do not possess molariform UPJ and LPJ dentition.

Regardless of standard length, sex, or sexual maturity, specimens of P. i t a s y are characterized by a markedly decurved predorsal profile owing to a pronounced occipital hump ( Figs. 2–3 View FIGURE 2 View FIGURE 3 ). Development of the occipital hump is less pronounced in P. betsileanus , in which only large males (> 180 mm SL) exhibit a prominent hump (de Rham & Nourissat 2002: 66, 68). A well­developed hump is lacking in female P. betsileanus , regardless of standard length (de Rham & Nourissat 2002: 68), and is either absent or poorly developed in males less than about 180 mm SL (de Rham & Nourissat 2002: 70).

Local names. ‘Trondro mainty’, which translates as black (= mainty) fish (= trondro), is the Malagasy name most commonly used to refer to P. i t a s y. The new species is also referred to as 'Marakely à bosse' in the literature.

Etymology. Named for the crater lake in the central highlands of Madagascar where all members of the type series were presumably collected. The epithet, itasy , is used as a noun in apposition.

TABLE 1. Morphometric and meristic data of Ptychochromoides betsileanus and P. itasy. Measurements (mm) in percent standard length (SL) or percent head length (HL), unless noted otherwise. Numbers in parentheses correspond to number of individuals with indicated count.

character Standard length (mm) Head length % SL n 10 10 P. betsileanus range mean 125.0–320.0 175.3 30.0–33.8 32.2 SD 1.09 n 4 3 holotype 123.2 31.1 P. itasy range 77.5–123.2 31.1–32.5 mean 99.6 31.6 SD 0.85
Head width (max.) % SL 10 17.2–21.7 18.8 1.43 3 18.2 18.2–18.5 18.4 0.18
Body depth % SL 10 45.5–52.5 47.8 1.92 3 46.4 45.2–46.4 45.7 0.52
Predorsal length % SL 10 40.4–46.2 42.7 1.69 3 39.3 39.3–40.2 39.6 0.4
Preanal length % SL Prepelvic length % SL Caudal peduncle length % SL Caudal peduncle width % SL Caudal peduncle depth % SL 10 10 9 10 10 69.8–75.4 72.3 39.5–42.7 41.0 12.9–18.8 15.3 5.8–9.5 7.6 16.2–17.5 17 1.57 0.86 1.76 1.32 0.36 3 3 3 3 3 71.9 39.8 15.2 7.3 17.2 71.9–73.3 39.8–43.6 14.6–15.2 6.8–7.4 15.8–17.5 72.6 41.2 15 7.1 16.8 0.54 1.72 0.3 0.26 0.75
Pectoral­fin length % SL 10 21.9–27.4 24.5 1.88 3 22.8 22.8–23.5 23.2 0.27
Pelvic­fin length % SL 10 23.9–29.6 25.5 1.57 3 27.6 27.4–27.6 27.5 0.09
Last dorsal spine length % SL 6 11.4–13.9 13 0.97 3 16.4 14.1–16.4 15.3 0.97
Snout length % HL 10 36.1–45.1 39.6 2.84 3 35.3 33.2–35.7 34.7 1.07
Orbit diameter % HL Upper jaw length % HL Lower jaw length % HL Interorbital width % HL Preorbital depth % HL 10 10 10 10 10 19.9–31.9 27.6 30.6–45.1 36.5 37.6–47.6 41.9 37.5–44.4 41.2 18.3–27.4 22.8 3.72 3.96 3.20 1.90 2.33 3 3 3 3 3 30.3 33.7 41.0 42.3 20.1 30.3–33.6 33.3–34.6 40.0–43.0 38.5–42.3 17.8–20.1 31.4 33.9 41.3 40.2 19.2 1.51 0.54 1.25 1.60 0.99
Caudal peduncle length/width 9 1.6–2.5 2.0 0.26 3 2.1 2.1–2.2 2.1 0.04
Caudal peduncle length/depth 9 0.8–1.1 0.9 0.1 3 0.9 0.8–1.0 0.9 0.05

TABLE 2. Loadings of morphometric variables in sheared principal components analysis of Ptychochromoides betsileanus, P. i t a s y, and Ptychochromoides sp. (MNHN 1907 ­ 104).

Variable sheared PC2 sheared PC3
Standard length ­0.037 0.053
Head length 0.097 0.041
Snout length ­0.072 0.117
Orbit diameter ­0.042 0.423
Upper jaw length 0.515 ­0.337
Lower jaw length 0.319 ­0.272
Interorbital width ­0.097 0.105
Preorbital depth ­0.722 ­0.413
Pectoral­fin length 0.118 0.341
Pelvic­fin length 0.198 ­0.071
Predorsal length ­0.124 0.452
Preanal length 0.025 0.297
Prepelvic length 0.125 0.147

University of Michigan, Museum of Zoology


American Museum of Natural History


Museum National d'Histoire Naturelle














Ptychochromoides itasy

Sparks, John S. 2004

Ptychochromoides betsileanus: Kiener and Maugé, 1966 : 77

Sparks 2001: 120
Reinthal 1997: 358
Kiener 1966: 77

Ptychochromis betsileanus:

Kiener 1963: 166
Arnoult 1959: 72
Pellegrin 1933: 144
GBIF Dataset (for parent article) Darwin Core Archive (for parent article) View in SIBiLS Plain XML RDF