Editorial Type: Articles
 | 
Online Publication Date: 18 Jul 2017

A New Species of Labeotropheus (Perciformes: Cichlidae) from Southern Lake Malaŵi, Africa

Article Category: Research Article
Page Range: 399 – 414
DOI: 10.1643/CI-16-463
Save
Download PDF

Labeotropheus is a genus of haplochromine cichlids endemic to Lake Malaŵi, Africa. Since its original description, surprisingly few additional species have been described, largely due to a misinterpretation of the historical taxonomic treatment of allopatric populations of Labeotropheus. Previously published evidence of genetic and morphological divergence among southern populations of Labeotropheus suggests that species diversity is underestimated. Morphometric, meristic, and color characteristics are employed to describe a new species of Labeotgropheus endemic to Mumbo Island and Thumbi West Island near the Nankumba Peninsula in southern Lake Malaŵi. The new species is distinguished from congeners by a wider body, longer pectoral fins, a shorter snout, and male nuptial coloration.

THE genus Labeotropheus is a small genus of haplochromine cichlids endemic to Lake Malaŵi that, until recently, contained only two species, L. fuelleborni (Ahl, 1927) and L. trewavasae (Fryer, 1956); currently, there are four species, with the recent inclusions of L. chlorosiglos and L. simoneae (Pauers, 2016). Labeotropheus is widely distributed throughout the lake and is distinguished from other Lake Malaŵi cichlids due to the presence of a fleshy pad present on the snout (Ahl, 1927; Oliver and Arnegard, 2010; Pauers, 2016). Despite having been somewhat overlooked by cichlid taxonomists (Pauers, 2010), this genus has long been a favorite research subject for evolutionary biologists. In particular, populations of Labeotropheus from the southern end of Lake Malaŵi, especially allopatric populations of the deep-bodied, shallow-dwelling form provisionally assigned to L. fuelleborni (Ribbink et al., 1983a, 1983b), have been the subject of many studies suggesting divergence among these populations.

One of the first studies to quantify genetic differentiation among southern populations of L. fuelleborni was that of Kornfield (1974). In this study, allozyme frequencies among these populations were examined, and the islands near the Nankumba Peninsula, including Mumbo Island and Thumbi West Island, were a particular focus. The populations of L. fuelleborni found at these islands had a high degree of similarity in allozyme expression, indicating a high likelihood of gene flow between these islands. In the late 1970s and early 1980s, Ribbink et al. (1983a, 1983b) performed several critical studies that greatly expanded what was known about the ecology and natural history of Labeotropheus spp. Among their many findings, Ribbink et al. (1983a, 1983b) found that the deep-bodied Labeotropheus present at Mumbo Island and Thumbi West Island shared a similar depth distribution (maximum depth 8–10 m); a similar substrate preference (the upper surfaces of large rocks); and a distinctive male nuptial color pattern consisting of a dark blue body and orange-red pigmentation on the pelvic fins, the anal fin, and the rayed portion of the dorsal fin. Later still, Arnegard et al. (1999), using microsatellite alleles, confirmed the genetic distinctiveness of the Mumbo Island population of L. fuelleborni, but did not examine the deep-bodied Labeotropheus found at Thumbi West Island.

Pauers and McMillan (2015) performed a geographically extensive analysis of morphological divergence among several populations of Labeotropheus (ten populations of L. fuelleborni, nine L. trewavasae) throughout Lake Malaŵi. While they found widespread differences in body shape among populations of both deep-bodied and slender-bodied Labeotropheus, they also found that the deep-bodied forms present at Mumbo Island and Thumbi West Island shared a similar, unique morphology (see Pauers and McMillan, 2015: morphotype b in figures 5 and 6, pp. 152–153). Inspired by this unique body shape, as well as the aforementioned ecological and genetic differences, I will describe the deep-bodied Labeotropheus found at these islands as a distinct species.

MATERIALS AND METHODS

Specimens of deep-bodied Labeotropheus from Mumbo Island and Thumbi West Island in southern Lake Malaŵi (Fig. 1) were obtained from several museum collections; all institutional abbreviations follow Sabaj (2016).

Fig. 1. . Map of Lake Malaŵi displaying where specimens were collected. (A) Lake Malaŵi. (B) Detail of Nankumba Peninsula, including Mumbo Island and Thumbi West Island.Fig. 1. . Map of Lake Malaŵi displaying where specimens were collected. (A) Lake Malaŵi. (B) Detail of Nankumba Peninsula, including Mumbo Island and Thumbi West Island.Fig. 1. . Map of Lake Malaŵi displaying where specimens were collected. (A) Lake Malaŵi. (B) Detail of Nankumba Peninsula, including Mumbo Island and Thumbi West Island.
Fig. 1.  Map of Lake Malaŵi displaying where specimens were collected. (A) Lake Malaŵi. (B) Detail of Nankumba Peninsula, including Mumbo Island and Thumbi West Island.

Citation: Ichthyology & Herpetology 105, 2; 10.1643/CI-16-463

The type series of L. fuelleborni was captured at Lumbila, Tanzania, and contains only five specimens. Similarly, the type series of L. trewavasae was obtained at Nkhata Bay, Malawi, and contains 12 specimens. In order to improve the reliability of comparisons between the type species and the putative new species, I used Pauers and McMillan (2015) to select populations with similar morphologies to these type series. In the case of L. fuelleborni, the population at Messule, Mozambique, is most similar to that of the type specimens; while somewhat distant from each other (∼227.4 km; Pauers and McMillan, 2015), both of these populations are found at locations along the eastern coast of Lake Malaŵi. For the type series of L. trewavasae, the most similarly shaped specimens come from the same island group at which the putative new taxon is found, the Mitande Rock/Mumbo Island/Thumbi West Island group; these are also somewhat distant (∼260 km) from Nkhata Bay. Therefore, the 23 specimens of L. fuelleborni from Messule, Mozambique, were added to the five type specimens from Lumbila, Tanzania (total n = 28), and the 11 specimens of L. trewavasae from the Mitande/Mumbo/Thumbi West group were added to the 12 type specimens from Nkhata Bay, Malaŵi (total n = 23). For both species, the additional specimens are quite similar with regards to meristic characteristics (Table 1), though there are some distinct morphometric differences between the types of L. trewavasae and the Mitande/Mumbo/Thumbi West specimens (e.g., dorsal-fin base length, dorsal-fin origin to anal-fin origin, body depth, preorbital depth, snout length, and lower jaw length; Table 2).

Table 1.  Meristic values of the three traditional species of Labeotropheus, and the Messule population of L. fuelleborni and the Mitande/Mumbo/Thumbi West population of L. trewavasae. Values found in holotypes (L. curvirostris and L. trewavasae) or lectotype (L. fuelleborni) are in bold following the modal value.

          Table 1. 
Table 2.  Morphometric values of the three traditional species of Labeotropheus, and the Messule population of L. fuelleborni and the Mitande/Mumbo/Thumbi West population of L. trewavasae. Values found in holotypes (L. curvirostris and L. trewavasae) or lectotype (L. fuelleborni) are in bold beneath the means. SE = Standard error.

          Table 2. 

External counts and measurements follow Barel et al. (1977), Stauffer et al. (1997), and Pauers (2016). Standard length is used throughout. Except for the counts of gill rakers, all counts and measurements were made on the left side of the fish. All measurements were taken to the nearest hundredth mm using digital calipers, and then rounded to the nearest one-tenth mm. Measurements taken on the trunk of the fishes were standardized by standard length, while those on the head were standardized by head length.

In order to diagnose putative new species of Labeotropheus, I followed Pauers (2016). Briefly, by first using the Evolutionary Species Concept (Simpson, 1961; Wiley, 1978) as a non-operational guiding principle (Mayden, 1999), the application of operational criteria was then used to distinguish one species from another. Within Labeotropheus, a useful first criterion to employ is the ratio of body depth to body length (Fryer, 1956; Pauers, 2016), as this is a useful first step in distinguishing deeper-bodied from slender bodied species (Pauers and McMillan, 2015), especially if they are sympatric with one another (sensu Stauffer and McKaye, 2001). Once a putative species has been shown to be more similar to L. fuelleborni, or more similar to L. trewavasae, or unique, further comparisons can be made to the appropriate type specimens utilizing other morphological, meristic, chromatic, or behavioral criteria.

In the case of overlap among morphometric and meristic characteristics, two types of factor analysis were used to assist in distinguishing species. Differences in body shape were examined by performing a sheared principal component analysis (SPCA) on the morphometric data. This analysis factored the covariance matrix and restricts size variation to the first principal component (Humphries et al., 1981; Bookstein et al., 1985). Meristic data were analyzed using principal component analysis (PCA) in which the correlation matrix was factored. Comparisons among species were illustrated by plotting pairs of morphometric principal components and meristic principal components (e.g., Stauffer et al., 1997).

Fig. 2. . Holotype of L. artatorostris, female, 114.9 mm SL, UMMZ 250296.Fig. 2. . Holotype of L. artatorostris, female, 114.9 mm SL, UMMZ 250296.Fig. 2. . Holotype of L. artatorostris, female, 114.9 mm SL, UMMZ 250296.
Fig. 2.  Holotype of L. artatorostris, female, 114.9 mm SL, UMMZ 250296.

Citation: Ichthyology & Herpetology 105, 2; 10.1643/CI-16-463

Fig. 3. . Graph of body depth versus standard length in all species of Labeotropheus. Labeotropheus artatorostris is clearly distinguished from the slender-bodied species, including L. simoneae; the type series of L. trewavasae; and the sympatric (i.e., Mitande/Mumbo/Thumbi West) form of the slender-bodied species currently assigned to L. trewavasae. The body depth-standard length ratio of L. artatorostris overlaps completely with that of L. fuelleborni; see Table 3 for details.Fig. 3. . Graph of body depth versus standard length in all species of Labeotropheus. Labeotropheus artatorostris is clearly distinguished from the slender-bodied species, including L. simoneae; the type series of L. trewavasae; and the sympatric (i.e., Mitande/Mumbo/Thumbi West) form of the slender-bodied species currently assigned to L. trewavasae. The body depth-standard length ratio of L. artatorostris overlaps completely with that of L. fuelleborni; see Table 3 for details.Fig. 3. . Graph of body depth versus standard length in all species of Labeotropheus. Labeotropheus artatorostris is clearly distinguished from the slender-bodied species, including L. simoneae; the type series of L. trewavasae; and the sympatric (i.e., Mitande/Mumbo/Thumbi West) form of the slender-bodied species currently assigned to L. trewavasae. The body depth-standard length ratio of L. artatorostris overlaps completely with that of L. fuelleborni; see Table 3 for details.
Fig. 3.  Graph of body depth versus standard length in all species of Labeotropheus. Labeotropheus artatorostris is clearly distinguished from the slender-bodied species, including L. simoneae; the type series of L. trewavasae; and the sympatric (i.e., Mitande/Mumbo/Thumbi West) form of the slender-bodied species currently assigned to L. trewavasae. The body depth-standard length ratio of L. artatorostris overlaps completely with that of L. fuelleborni; see Table 3 for details.

Citation: Ichthyology & Herpetology 105, 2; 10.1643/CI-16-463

Table 3.  Body depth-standard length ratio among Labeotropheus artatorostris, L. chlorosiglos, L. fuelleborni (type series + Messule specimens), L. trewavasae (both the type series and those sympatric with L. artatorostris), and L. simoneae.

          Table 3. 
Fig. 4. . Graph of snout pad length versus head length in Labeotropheus. (A) All species; (B) L. artatorostris and L. fuelleborni (Lumbila + Messule). Labeotropheus artatorostris has a distinctly shorter snout pad than the type specimens of L. fuelleborni, though there is some overlap between L. artatorostris and the Messule population of L. fuelleborni. Nonetheless, the differences between the snout pad lengths of L. artatorostris and both populations of L. fuelleborni are statistically significant; see Table 4 for details.Fig. 4. . Graph of snout pad length versus head length in Labeotropheus. (A) All species; (B) L. artatorostris and L. fuelleborni (Lumbila + Messule). Labeotropheus artatorostris has a distinctly shorter snout pad than the type specimens of L. fuelleborni, though there is some overlap between L. artatorostris and the Messule population of L. fuelleborni. Nonetheless, the differences between the snout pad lengths of L. artatorostris and both populations of L. fuelleborni are statistically significant; see Table 4 for details.Fig. 4. . Graph of snout pad length versus head length in Labeotropheus. (A) All species; (B) L. artatorostris and L. fuelleborni (Lumbila + Messule). Labeotropheus artatorostris has a distinctly shorter snout pad than the type specimens of L. fuelleborni, though there is some overlap between L. artatorostris and the Messule population of L. fuelleborni. Nonetheless, the differences between the snout pad lengths of L. artatorostris and both populations of L. fuelleborni are statistically significant; see Table 4 for details.
Fig. 4.  Graph of snout pad length versus head length in Labeotropheus. (A) All species; (B) L. artatorostris and L. fuelleborni (Lumbila + Messule). Labeotropheus artatorostris has a distinctly shorter snout pad than the type specimens of L. fuelleborni, though there is some overlap between L. artatorostris and the Messule population of L. fuelleborni. Nonetheless, the differences between the snout pad lengths of L. artatorostris and both populations of L. fuelleborni are statistically significant; see Table 4 for details.

Citation: Ichthyology & Herpetology 105, 2; 10.1643/CI-16-463

Table 4.  Snout pad length-head length ratio in Labeotropheus artatorostris and L. fuelleborni (both the Lumbila and Messule populations). Both snout pad and head length were log-transformed before analysis.

          Table 4. 
Fig. 5. . Plots of meristic PC 1 versus sheared morphometric PC 2 for the three deep-bodied species of Labeotropheus (L. artatorostris, L. chlorosiglos, and L. fuelleborni). (A) Labeotropheus artatorostris, L. chlorosiglos, and the expanded set of specimens of L. fuelleborni (Lumbila and Messule); (B) L. artatorostris, L. chlorosiglos, and the type specimens of L. fuelleborni. See Tables 5 and 6 for the meristic and morphometric loadings, respectively.Fig. 5. . Plots of meristic PC 1 versus sheared morphometric PC 2 for the three deep-bodied species of Labeotropheus (L. artatorostris, L. chlorosiglos, and L. fuelleborni). (A) Labeotropheus artatorostris, L. chlorosiglos, and the expanded set of specimens of L. fuelleborni (Lumbila and Messule); (B) L. artatorostris, L. chlorosiglos, and the type specimens of L. fuelleborni. See Tables 5 and 6 for the meristic and morphometric loadings, respectively.Fig. 5. . Plots of meristic PC 1 versus sheared morphometric PC 2 for the three deep-bodied species of Labeotropheus (L. artatorostris, L. chlorosiglos, and L. fuelleborni). (A) Labeotropheus artatorostris, L. chlorosiglos, and the expanded set of specimens of L. fuelleborni (Lumbila and Messule); (B) L. artatorostris, L. chlorosiglos, and the type specimens of L. fuelleborni. See Tables 5 and 6 for the meristic and morphometric loadings, respectively.
Fig. 5.  Plots of meristic PC 1 versus sheared morphometric PC 2 for the three deep-bodied species of Labeotropheus (L. artatorostris, L. chlorosiglos, and L. fuelleborni). (A) Labeotropheus artatorostris, L. chlorosiglos, and the expanded set of specimens of L. fuelleborni (Lumbila and Messule); (B) L. artatorostris, L. chlorosiglos, and the type specimens of L. fuelleborni. See Tables 5 and 6 for the meristic and morphometric loadings, respectively.

Citation: Ichthyology & Herpetology 105, 2; 10.1643/CI-16-463

Table 5.  Meristic values of species of Labeotropheus, including L. artatorostris, new species. Values found in holotypes (L. artatorostris, L. chlorosiglos, L. curvirostris, L. simoneae, and L. trewavasae,) or lectotype (L. fuelleborni) are in bold following the modal value.

          Table 5. 
Table 6.  Morphometric data for all species of Labeotropheus, including L. artatorostris, new species. Values for holotypes (L. artatorostris, L. chlorosiglos, L. curvirostris, L. simoneae, and L. trewavasae) or paratype (L. fuelleborni) are in bold beneath the means. SE = Standard error.

          Table 6. 
Fig. 6. . Typical color patterns of adult male L. artatorostris. (A) Non-type, wild male photographed at Thumbi West Island; (B) non-type, wild males engaged in agonistic behavior, photographed at Thumbi West Island; (C) non-type, wild male photographed at Mumbo Island. All photographs were generously provided by, and are © Ad Konings/Cichlid Press.Fig. 6. . Typical color patterns of adult male L. artatorostris. (A) Non-type, wild male photographed at Thumbi West Island; (B) non-type, wild males engaged in agonistic behavior, photographed at Thumbi West Island; (C) non-type, wild male photographed at Mumbo Island. All photographs were generously provided by, and are © Ad Konings/Cichlid Press.Fig. 6. . Typical color patterns of adult male L. artatorostris. (A) Non-type, wild male photographed at Thumbi West Island; (B) non-type, wild males engaged in agonistic behavior, photographed at Thumbi West Island; (C) non-type, wild male photographed at Mumbo Island. All photographs were generously provided by, and are © Ad Konings/Cichlid Press.
Fig. 6.  Typical color patterns of adult male L. artatorostris. (A) Non-type, wild male photographed at Thumbi West Island; (B) non-type, wild males engaged in agonistic behavior, photographed at Thumbi West Island; (C) non-type, wild male photographed at Mumbo Island. All photographs were generously provided by, and are © Ad Konings/Cichlid Press.

Citation: Ichthyology & Herpetology 105, 2; 10.1643/CI-16-463

Table 7.  Analysis of principal components (sheared morphometric PC 2 and meristic PC 1) among Labeotropheus sp. Tables 7A and 7B (corresponding to Figure 5A): Labeotropheus fuelleborni Lumbila (type series) + Messule, L. chlorosiglos, and L. artatorostris; Tables 7C and 7D (corresponding to Figure 5B): L. fuelleborni Lumbila (type series), L. chlorosiglos, and L. artatorostris.
Table 7. 
Table 8.  Sheared morphometric principal component loadings. PCs 2–6 are sheared. Table 8A (corresponding to Figure 5A): Labeotropheus fuelleborni Lumbila (type series) + Messule, L. chlorosiglos, and L. artatorostris; 8B (corresponding to Figure 5B): L. fuelleborni Lumbila (type series), L. chlorosiglos, and L. artatorostris.

          Table 8. 
Table 9.  Meristic principal component loadings. Table 9A (corresponding to Figure 5A): Labeotropheus fuelleborni Lumbila (type series) + Messule, L. chlorosiglos, and L. artatorostris; 9B (corresponding to Figure 5B): L. fuelleborni Lumbila (type series), L. chlorosiglos, and L. artatorostris.

          Table 9. 

Holotype

UMMZ 250296 (Field No. PNR85-MUM-5-25), adult female, 114.9 mm SL, Mumbo Island, Lake Malaŵi, Africa, Peter N. Reinthal, 25 December 1985 (Fig. 2).

Paratypes

AMNH 92810 (Field No. PNR-89-25), 3 males (79.0, 83.5, and 87.7 mm SL), 1 female (75.6 mm SL), Africa, Lake Malaŵi, Thumbi West Island, Peter N. Reinthal, 6 June 1989; MCZ 157254, 1 female, 71.6 mm SL, USNM 261886 (Field No. MKO 80-7), 2 males (101.7, 107.1 mm SL), 1 female (71.4 mm SL), Malawi, Lake Malaŵi, Mumbo Island, along W shore of cove on N side (cove is ca. 50 m wide, 100 m long, U-shaped, and is pointed directly at mtns of peninsula at Salima), K. R. McKaye et al., 3 June 1980; MPM Fi50050 (Field No. PNR85-MUM-5-25), 1 male (87.7 mm SL), 1 female (98.0 mm SL), UMMZ 238339 (Field No. PNR85-MUM-5-25), 3 males (83.5, 99.7, and 101.2 mm SL), 8 females (73.9, 79.9, 83.6, 86.9, 88.5, 98.4, 104.3, and 114.0 mm SL), Africa, Lake Malaŵi, Mumbo Island, Peter N. Reinthal, 25 December 1985; USNM 261919 (Field No. MKO 80-21), 2 males (66.2, 80.1 mm SL), 6 females (54.4, 62.1, 66.1, 70.0, 70.3, and 74.0 mm SL), Malawi, Lake Malaŵi, Thumbi I. West, bay facing N at W tip of N shore, from offshore rocks to ca. midway along W shore of bay, K. R. McKaye et al., 12 June 1980.

Non-type material

MCZ 172136, 1 unmeasured and dissected (brain removed) juvenile, MCZ 172137, 1 unmeasured juvenile, USNM 438972 (ex-USNM 261886, Field No. MKO 80-7), 6 unmeasured juveniles (≤40.0 mm SL), Malawi, Lake Malaŵi, Mumbo Island, along W shore of cove on N side (cove is ca. 50 m wide, 100 m long, U-shaped, and is pointed directly at mtns of peninsula at Salima), K. R. McKaye et al., 3 June 1980; USNM 438973 (ex-USNM 261919, Field No. MKO 80-21), 1 unmeasured juvenile, Malawi, Lake Malaŵi, Thumbi I. West, bay facing N at W tip of N shore, from offshore rocks to ca. midway along W shore of bay, K. R. McKaye et al., 12 June 1980.

Diagnosis

All diagnostic characteristics of Labeotropheus are present, including a steeply sloping head; broad, fleshy snout; wide jaws with retrognathous lower jaw; and an inferior and subterminal mouth. Labeotropheus artatorostris differs from the slender-bodied Labeotropheus, L. trewavasae Fryer and L. simoneae Pauers, by its deeper body depth (34.3–42.0% SL vs. 26.9–30.8% in L. simoneae, and 25.3–33.4% in all specimens of L. trewavasae; Fig. 3; Table 3), greater distance between insertion of dorsal fin and origin of anal fin (29.2–34.0% SL vs. 27.4–28.6% in L. simoneae and 26.3–29.1% in all specimens of L. trewavasae), and greater distance between origin of dorsal fin and insertion of pelvic fins (36.1–41.1% SL vs. 27.5–32.8% in L. simoneae, 26.1–32.7% in all specimens of L. trewavasae). Labeotropheus artatorostris is different from L. curvirostris by having more teeth rows in both the upper and lower jaws (5–8 vs. 3 and 3–8 vs. 1, respectively), more infraorbital neuromasts (12–36 vs. 9), a longer pectoral fin (24.6–33.3% SL vs. 24.1%), shorter upper jaw (15.6–22.9% HL vs. 23.6%), and a narrower snout (33.2–40.2% HL vs. 42.6%). It differs from L. fuelleborni due to a shorter snout pad (7.43–14.2% HL vs. 14.9–17.2% in L. fuelleborni and 9.0–17.2% in types of L. fuelleborni + Messule specimens; Fig. 4; Table 2), more rows of teeth in the upper jaw (5–8 vs. 4–5 in L. fuelleborni and 4–7 in L. fuelleborni types + Messule specimens), and more gillrakers on the first ceratobranchial (7–10 vs. 7–8 in L. fuelleborni). It differs from L. chlorosiglos Pauers due to greater distance between the opercular tabs (16.6–18.7% SL vs. 14.7–15.7% in L. chlorosiglos), and greater distance between origin of dorsal fin and insertion of pelvic fin (36.1–41.1% SL vs. 33.0–36.0% in L. chlorosiglos). Labeotropheus artatorostris also has longer pectoral fins (24.6–33.3% SL) than all other species of Labeotropheus: L. chlorosiglos (22.2–26.0% SL), L. fuelleborni (23.1–25.7% SL), L. fuelleborni types + Messule specimens (22.2–29.3% SL), L. simoneae (21.9–26.3% SL), and all specimens of L. trewavasae (18.4–23.9% SL). The new species also has a wider body than all other species of Labeotropheus: body width at pectoral fins 14.0–18.3% SL vs. 12.7–14.3% in L. chlorosiglos, 15.0–16.9% in L. fuelleborni (and 13.1–17.9% in L. fuelleborni types + Messule specimens), 13.0–15.6% in L. simoneae, and 12.1–16.6% in L. trewavasae; body width at opercular tabs 16.6–18.7% SL vs. 14.7–15.7% in L. chlorosiglos, 16.2–18.2% in L. fuelleborni (and 15.3–18.4% in types of L. fuelleborni + Messule specimens), 14.7–16.7% in L. simoneae, and 12.1–16.6% in all specimens of L. trewavasae. Finally, L. artatorostris differs from all other species of Labeotropheus based on fin coloration of males: presence of orange-red pigmentation in pelvic, anal, and rayed dorsal fins in L. artatorostris vs. all fins blue in L. fuelleborni and L. trewavasae, and whitish-blue anal and yellow-rayed dorsal fins in L. chlorosiglos and L. simoneae. Labeotropheus artatorostris also lacks the large orange flank patch found in L. chlorosiglos and L. simoneae, and the orange pigmentation present on operculum of L. simoneae.

Description

Meristic data in Table 5, morphometric data in Table 6. Contrary to Ahl's (1927) description of the genus, all oral teeth tricuspid, and all scales ctenoid, except belly and breast, which are cycloid.

Body depth 34.3–42.0% of SL, and distance between dorsal-fin insertion and anal-fin insertion 15.4–18.3% of SL. Body wide; width at opercular tabs 16.6–18.7% HL and width at pectoral fins 14.0–18.3% HL. Anterior lateral line with 21–24 scales, posterior lateral line with 9–16 scales, and 0–6 scales overlapping between the anterior and posterior lateral lines. Dorsal fin 56.2–64.5% SL with 15–18 spines and 8–9 rays. Pectoral fins long (24.6–33.3% SL), rounded, with 12–15 rays. Pelvic fins long, especially in mature males, reaching anal-fin origin and sometimes beyond.

Head long (head length 29.9–34.9% of SL) and relatively deep (head depth 93.6–114.8% of HL); cheek with 2–5 scale rows. Snout relatively long but narrow; snout length 25.2–40.8% HL, snout width 33.3–40.2% HL. Snout pad short, 7.4–14.2% HL. Oral jaws compact: lower jaws long but narrow, 23.7–39.3% of HL and 39.0–48.1% of HL, respectively; but lateral length of mandible relatively long, 15.6–22.9% HL. Upper jaw with 5–8 rows of teeth, and 4–10 teeth on lateral portion of left upper jaw. Lower jaw 3–8 rows of teeth, and 19–35 teeth present in left half of first row of lower jaw. Gill rakers stout, triangular, and widely spaced; 7–10 on ceratobranchial and 1–4 epibranchial gill rakers on first gill arch. All specimens with 1 raker between the cerato- and epibranchial rakers.

Principal component analyses as described above were performed to compare L. artatorostris to L. fuelleborni. When the second sheared morphometric principal components are plotted against the second meristic principal component, two trends clearly emerge. First, both known populations of L. artatorostris, from Mumbo Island and Thumbi West Island, are very similar in morphometric and meristic characteristics, and overlap extensively on the factor plot. Labeotropheus artatorostris is distinct from the L. fuelleborni along the morphometric axis and is distinct from L. chlorosiglos along the meristic axis (Fig. 5, Table 7), and these differences hold whether the expanded sample of L. fuelleborni (Lumbila + Messule) or only the types are used in the analysis. The loadings resulting from the sheared morphometric PCA are listed in Table 8, while those of the meristic PCA are found in Table 9.

Coloration of males

Coloration of living male L. artatorostris was extensively described in Ribbink et al. (1983a); coloration was described for both the Mumbo Island and Thumbi West populations, and was observed directly in Lake Malaŵi:

“Thumbi West Island: Body and head blue. Dorsal fin blue with orange trailing edge and yellow ocelli, but less black in the rayed portion than Monkey Bay form. Anal fin blue at base, otherwise orange-pink with 3–8 yellow egg-dummies. Pelvic fins mainly orange with narrow black submarginal bands and whitish-blue leading edges.

Mumbo Island: Body and head dark blue. Dorsal fin pale blue with orange-red rayed region. Caudal fin rays blue with orange inter-ray membranes and yellow trailing edge. Anal fin orange with yellow egg-dummies. Pelvic fins orange-red with light blue leading edges...” (p. 239)

Photographs of L. artatorostris taken in Lake Malaŵi published in Konings (1995, 2001, 2007) and Spreinat (1997) largely agree with these descriptions, with the following exceptions: 11 dark bluish-black vertical bars occasionally visible across flank and caudal peduncle; two of these are present on caudal peduncle, nine across flank to head. Opercular tab bright metallic blue. Typical male coloration is shown in Figure 6.

In preservative, males uniform brown. Eleven vertical bars on caudal peduncle and flank visible on some specimens.

Coloration of females

Ribbink et al. (1983a) briefly described the coloration of living female L. artatorostris observed directly at Mumbo Island: “Females dark blue-grey and OB (author's note: ‘OB' refers to a color pattern known as “orange-blotch,” common in rock-dwelling cichlids from Lake Malaŵi, that consists of small orange, black, white, gray, and bluish blotches scattered over the body of the fish), but the OB form is rare” (p. 239). Published photographs of female L. artatorostris exceedingly rare, though Ribbink et al. (1983a) do describe likely color for female Labeotropheus.

In preservative, females uniform brown. Eleven vertical bars on caudal peduncle and flank visible on some specimens.

Distribution

Labeotropheus artatorostris is endemic to Lake Malaŵi, and seems restricted to Mumbo Island and Thumbi West Island, although Arnegard et al. (1999) report deep-bodied Labeotropheus possessing similar male nuptial coloration along the eastern coast of the Nankumba Peninsula from Songwe Hill northwest to the Ilala Gap, including Zimbawe Rock. Additionally, it may possibly be found at the Mitande Rocks on the eastern end of Thumbi West Island, but specimens from Mitande are rare in museum collections. Further fieldwork in Lake Malaŵi is necessary to determine the full distribution of L. artatorostris.

Genetics

Kornfield (1974) found that the Mumbo Island and Thumbi West Island populations of what was then identified as L. fuelleborni shared alleles of the esterase-1 (e.g., EST-195) and phosphohexose isomerase-3 (e.g., PHI-31.00/1.00) allozymes, and together were genetically distinct from other populations of L. fuelleborni. Similarly, Arnegard et al. (1999) found that the Mumbo Island population of what was then identified as L. fuelleborni had unique alleles present at the UNH001, UNH002, UNH050, and UNH231 microsatellite loci.

Etymology

The specific epithet is a composite of two Latin words, artatus, meaning constricted or shortened, and rostris, meaning nose or snout. This refers to the short snout pad that is characteristic of this species, especially in comparison to the type specimens of L. fuelleborni.

DISCUSSION

The description of Labeotropheus artatorostris raises the number of species of Labeotropheus to five. As reported in Pauers (2010) and Pauers (2016), many biologists argued strenuously against the existence of species of Labeotropheus other than L. fuelleborni and L. trewavasae. This was largely due to a misunderstanding of the decision of Ribbink et al. (1983a, 1983b) not to recognize allopatric populations of both L. fuelleborni and L. trewavasae as separate species, which was done to temporarily mitigate the taxonomic uncertainty surrounding these allopatric populations. Ribbink et al. (1983a, 1983b) also suggested that while gene flow among these populations was highly unlikely, the presence of distinct male nuptial colors within each population made reproductive isolation among populations of Labeotropheus highly probable. The presence of both unique allozyme (Kornfield, 1974) and microsatellite alleles (Arnegard et al., 1999), and a distinctive male nuptial color pattern supports the species-status of Labeotropheus artatorostris (Ribbink et al., 1983a, 1983b; Pauers, 2016).

The description of Labeotropheus artatorostris, as well as those of L. chlorosiglos and L. simoneae, underscores the surprisingly limited state of our current knowledge of the diversity of the Labeotropheus. Despite decades of study of the evolution, speciation, and divergence of cichlids from Lake Malaŵi, this genus, in particular, has suffered from a lack of focused attention, as well as from the aforementioned misunderstandings regarding the taxonomic standing of allopatric populations. It is crucial, then, to conduct further investigations of the taxonomy of the Labeotropheus to better ascertain the distribution and diversity of species within this genus.

MATERIAL EXAMINED

Labeotropheus chlorosiglos

MPM50035 (1 alc. paratype), 114.3 mm SL, MPM50039 (1 alc. holotype), 103.5 mm SL, MPM50040 (7 alc. paratypes), 89.0–101.0 mm SL, Katale Island, Lake Malaŵi, collected by Stuart M. Grant, Ltd.

Labeotropheus fuelleborni

MRAC 79-40-P-89-110 (15 adult alc. specimens + 7 unmeasured juveniles), 56.7–104.3 mm SL, Mozambique, Messule, collected by Brichard; MRAC 99-041-P-1610-1612 (3 alc. specimens), 85.7–88.8 mm SL, MRAC 99-041-P-1635-1639 (5 alc. specimens), 71.8–94.4 mm SL, Mozambique, Messule, collected by SADEC/GEF; ZMB 22707 (1 alc. lectotype), 91.11 mm SL, ZMB 23922 (1 alc. paralectotype), 98.25 mm SL, ZMB 23923 (1 alc. paralectotype), 65.72 mm SL, ZMB 33818 (1 alc. paralectotype), 87.88 mm SL, ZMB 33819 (1 alc. paralectotype), 100.82 mm SL, Alt Langenburg (Lumbila, Tanzania), Lake Malaŵi, collected by Fuelleborn.

Labeotropheus simoneae

MPM 50048 (1 alc. holotype), 101.3 mm SL, MPM 50049 (4 alc. paratypes), 92.9–110.4 mm SL, submerged reef near Katale Island, Lake Malaŵi, collected by Stuart M. Grant, Ltd.

Labeotropheus trewavasae

BMNH 1965.11.2.1 (1 alc. holotype), 79.4 mm SL, BMNH 1965.11.2.2–12 (11 alc. paratypes), 61.04–89.45 mm SL, Nkhata Bay, Lake Malaŵi, collected by Fryer; MCZ 98593 (2 adult alc. specimens), 48.4–62.8 mm SL, Mumbo Island, Lake Malaŵi, collected by M. K. Oliver; UMMZ 237728 (1 adult alc. specimen), 70.0 mm SL, Thumbi West Island, Lake Malaŵi, collected by P. N. Reinthal; USNM 261913 (3 adult alc. specimens), 47.9–54.7 mm SL, Mumbo Island, Lake Malaŵi, collected by K. R. McKaye et al.; USNM 270447 (5 adult alc. specimens), 61.3–87.6 mm SL, Mitande Rocks, Lake Malaŵi, collected by P. N. Reinthal.

ACKNOWLEDGMENTS

Loans of specimens were arranged by B. Brown (AMNH), J. MacLaine (BMNH), K. Hartel (MCZ), D. Nelson (UMMZ), and J. Williams (USNM); D. Nelson was particularly gracious with his assistance and generosity regarding the type specimens for L. artatorostris. P. Bartsch and his staff, especially C. Lamour and J. Kapp, were gracious and generous hosts during my visit to ZMB to examine the type series of Labeotropheus, and I thank them for their kindness and hospitality. The color photos in Figure 6 were generously provided by A. Konings (Cichlid Press), and I am grateful for his kind and immediate attention to my request for help. I thank R. Henderson (MPM) for his assistance and suggestions for a specific epithet; additionally, S. Borkin, E. Censky, and J. Colby provided office space and administrative assistance at MPM. This work was supported by funds provided by the Orth Family Ichthyology Research Endowment at the Milwaukee Public Museum, UW Colleges Department of Biological Sciences Professional Development Funds, and a UW Colleges Summer Research Grant.

LITERATURE CITED

  • Ahl, E.
    1927. Einige neue Fische der Familie Cichlidae aus dem Nyassa-see. Sitzungsberichte der Berlinische Gesellschaft Naturforschender Freunde zu Berlin1926:5162.
  • Arnegard, M. E.,
    J. A. Markert,
    P. D. Danley,
    J. R. Stauffer, Jr.,
    A. J. Ambali,
    and
    T. D. Kocher.
    1999. Population structure and colour variation of the cichlid fish Labeotropheus fuelleborni Ahl along a recently formed archipelago of rocky habitat patches in southern Lake Malawi. Proceedings of the Royal Society, London (Series B)266:119130.
  • Barel, C.,
    M. van Oijen,
    F. Witte,
    and
    E. Witte-Maas.
    1977. An introduction to the taxonomy and morphology of the haplochromine Cichlidae from Lake Victoria. Netherlands Journal of Zoology27:333389.
  • Bookstein, F.,
    B. Chernoff,
    R. Elder,
    J. Humphries,
    G. Smith,
    and
    R. Strauss.
    1985. Morphometrics in Evolutionary Biology.
    Academy of Natural Sciences
    , Special Publication 15,
    Philadelphia
    .
  • Fryer, G.
    1956. A new species of Labeotropheus from Lake Nyasa, with a redescription of Labeotropheus fuelleborni Ahl and some notes on the genus Labeotropheus (Pisces: Cichlidae). Revue de Zoologie et de Botanique Africaines54:280289.
  • Humphries, J.,
    F. Bookstein,
    B. Chernoff,
    G. Smith,
    R. Elder,
    and
    S. Poss.
    1981. Multivariate discrimination by shape in relation to size. Systematic Zoology30:291308.
  • Konings, A.
    1995. Malawi Cichlids in Their Natural Habitat. Second edition.
    Cichlid Press
    ,
    El Paso, Texas
    .
  • Konings, A.
    2001. Malawi Cichlids in Their Natural Habitat. Third edition.
    Cichlid Press
    ,
    El Paso, Texas
    .
  • Konings, A.
    2007. Malawi Cichlids in Their Natural Habitat. Fourth edition.
    Cichlid Press
    ,
    El Paso, Texas
    .
  • Kornfield, I. L.
    1974. Evolutionary genetics of endemic cichlid fishes (Pisces: Cichlidae) in Lake Malawi, Africa. Unpubl. Ph.D. diss.,
    State University of New York
    ,
    Stony Brook
    .
  • Mayden, R. L.
    1999. Consilience and a hierarchy of species concepts: advances toward closure on the species puzzle. Journal of Nematology31:95116.
  • Oliver, M. K.,
    and
    M. E. Arnegard.
    2010. A new genus for Melanochromis labrosus, a problematic Lake Malawi cichlid with hypertrophied lips (Teleostei: Cichlidae). Ichthyological Exploration of Freshwaters21:209232.
  • Pauers, M. J.
    2010. Species concepts, speciation, and taxonomic change in the Lake Malawi mbuna, with special reference to the genus Labeotropheus Ahl 1927 (Perciformes: Cichlidae). Reviews in Fish Biology and Fisheries20:187202.
  • Pauers, M. J.
    2016. Two new and remarkably similarly colored species of Labeotropheus (Perciformes: Cichlidae) from Lake Malaŵi, Africa. Copeia104:628638.
  • Pauers, M. J.,
    and
    S. A. McMillan.
    2015. Geometric morphometrics reveals surprising diversity in the Lake Malawi cichlid genus Labeotropheus. Hydrobiologia748:145160.
  • Ribbink, A. J.,
    B. A. Marsh,
    A. C. Marsh,
    A. C. Ribbink,
    and
    B. J. Sharp.
    1983 a. A preliminary survey of the cichlid fishes of rocky habitats in Lake Malawi. South African Journal of Zoology18:149310.
  • Ribbink, A. J.,
    A. C. Marsh,
    B. A. Marsh,
    and
    B. J. Sharp.
    1983 b. The zoogeography, ecology and taxonomy of the genus Labeotropheus Ahl, 1927, of Lake Malawi (Pisces: Cichlidae). Zoological Journal of the Linnean Society79:223243.
  • Sabaj, M. H.
    (Ed.). 2016. Standard symbolic codes for institutional resource collections in herpetology and ichthyology: an Online Reference. Version 6.5 (16 August 2016). Electronically accessible at http://www.asih.org/, American Society of Ichthyologists and Herpetologists, Washington, D.C.
  • Simpson, G. G.
    1961. Principles of Animal Taxonomy.
    Columbia University Press
    ,
    New York
    .
  • Spreinat, A.
    1997. Aqualex-catalog: Cichlids from Lake Malawi.
    Dähne Verlag
    ,
    Ettlingen, Germany
    .
  • Stauffer, J. R., Jr.,
    N. J. Bowers,
    K. A. Kellogg,
    and
    K. R. McKaye.
    1997. A revision of the blue-black Pseudotropheus zebra (Teleostei: Cichlidae) complex from Lake Malawi, Africa, with a description of a new genus and ten new species. Proceedings of the Academy of Natural Sciences of Philadelphia148:189230.
  • Stauffer, J. R., Jr.,
    and
    K. A. McKaye.
    2001. The naming of cichlids. Journal of Aquariculture and Aquatic Sciences9:116.
  • Wiley, E. O.
    1978. The evolutionary species concept reconsidered. Systematic Zoology27:1726.
Copyright: © 2017 by the American Society of Ichthyologists and Herpetologists 2017
<bold>Fig. 1. </bold>
Fig. 1. 

Map of Lake Malaŵi displaying where specimens were collected. (A) Lake Malaŵi. (B) Detail of Nankumba Peninsula, including Mumbo Island and Thumbi West Island.


<bold>Fig. 2. </bold>
Fig. 2. 

Holotype of L. artatorostris, female, 114.9 mm SL, UMMZ 250296.


<bold>Fig. 3. </bold>
Fig. 3. 

Graph of body depth versus standard length in all species of Labeotropheus. Labeotropheus artatorostris is clearly distinguished from the slender-bodied species, including L. simoneae; the type series of L. trewavasae; and the sympatric (i.e., Mitande/Mumbo/Thumbi West) form of the slender-bodied species currently assigned to L. trewavasae. The body depth-standard length ratio of L. artatorostris overlaps completely with that of L. fuelleborni; see Table 3 for details.


<bold>Fig. 4. </bold>
Fig. 4. 

Graph of snout pad length versus head length in Labeotropheus. (A) All species; (B) L. artatorostris and L. fuelleborni (Lumbila + Messule). Labeotropheus artatorostris has a distinctly shorter snout pad than the type specimens of L. fuelleborni, though there is some overlap between L. artatorostris and the Messule population of L. fuelleborni. Nonetheless, the differences between the snout pad lengths of L. artatorostris and both populations of L. fuelleborni are statistically significant; see Table 4 for details.


<bold>Fig. 5. </bold>
Fig. 5. 

Plots of meristic PC 1 versus sheared morphometric PC 2 for the three deep-bodied species of Labeotropheus (L. artatorostris, L. chlorosiglos, and L. fuelleborni). (A) Labeotropheus artatorostris, L. chlorosiglos, and the expanded set of specimens of L. fuelleborni (Lumbila and Messule); (B) L. artatorostris, L. chlorosiglos, and the type specimens of L. fuelleborni. See Tables 5 and 6 for the meristic and morphometric loadings, respectively.


<bold>Fig. 6. </bold>
Fig. 6. 

Typical color patterns of adult male L. artatorostris. (A) Non-type, wild male photographed at Thumbi West Island; (B) non-type, wild males engaged in agonistic behavior, photographed at Thumbi West Island; (C) non-type, wild male photographed at Mumbo Island. All photographs were generously provided by, and are © Ad Konings/Cichlid Press.


Contributor Notes

Associate Editor: D. Buth.

Received: 29 May 2016
Accepted: 05 May 2017
  • Download PDF