ArticlePDF Available

Aphyosemion musafirii (Cyprinodontiformes: Nothobranchiidae), a new species from the Tshopo Province in the Democratic Republic of Congo, with some notes on the Aphyosemion of the Congo Basin.

The genus Aphyosemion was erected by Myers in 1924
with the type species A. castaneum, described in the same
publication, from Kisangani (Democratic Republic of
Congo). At present the taxonomy of the genus is still not
settled, here we use Aphyosemion as proposed in two re-
cent publications of the authors (Sonnenberg, 2007; Van
der Zee & Sonnenberg, 2010). This is identical with the
subgenus Aphyosemion of other authors (e.g. Collier,
2007; Huber, 2007; Murphy & Collier, 1999; Wildekamp,
1993) and consists of 16 species currently accepted as
valid, which are, with the exception of two species from
Gabon, endemic to the Congo drainage.
Only three species are currently known to occur in the
eastern part of the Congo Basin: A. christyi (Boulenger,
1915), A. schoutedeni (Boulenger, 1920), and A. casta-
neum Myers, 1924 (Fig. 1). A fourth species, A. margare-
tae Fowler, 1936, is currently considered as a synonym
to A. christyi (Van der Zee & Huber, 2006). The majori-
ty of museum collections of these species originate from
the right bank of the Congo River. Aphyosemion schout-
edeni was assumed to be restricted to the type locality
“Medje at the Naya River”, a tributary to the Aruwimi
Basin, about 300 km northeast of Kisangani. Although the
types are in good condition, all colour has disappeared.
Since nothobranchiid species, at least within species
groups or genera, differ little in morphological characters
(Scheel, 1968, 1990), colour pattern of the male is cru-
cial for species identification. Topotypes collected by Lang
and Chapin in 1910, however, still have their colour pat-
tern preserved (Van der Zee & Huber, 2006) and it is close
to that of A. castaneum with the exception of the anal fin
colour pattern. This colour pattern is also present in sev-
eral Aphyosemion collections in the Royal Museum for
Central Africa (MRAC) (Tervuren, Belgium) originating
from the Aruwimi Basin, east of the Kisangani-Buta road
Aphyosemion musafirii (Cyprinodontiformes: Nothobranchiidae),
a new species from the Tshopo Province in the Democratic Republic
of Congo, with some notes on the Aphyosemion of the Congo Basin
Jouke R. Van der Zee1& Rainer Sonnenberg2,3
1Royal Museum for Central Africa, Zoology Department, Ichthyology, Leuvensesteenweg 13,
B-3080 Tervuren, Belgium. Corresponding author. E-mail:
2Max-Planck-Institut für Evolutionsbiologie, August-Thienemann-Strasse 2, D-24306 Plön, Germany
(current address)
3Zoologisches Forschungsmuseum Alexander Koenig, Adenauerallee 160, D-53113 Bonn, Germany
Abstract. Aphyosemion musafirii, new species, is described from specimens collected near Ubundu (Ruiki River, Con-
go Basin, Tshopo Province, Democratic Republic of the Congo). Another population of A. musafirii is known from the
Romée River, 50 km West of Kisangani. The Ruiki and Romée Rivers are small tributaries on the left bank of the Con-
go River. Aphyosemion musafirii can be distinguished from its closest relative A. castaneum by the male colour pattern.
A preliminary DNA analysis demonstrates that Aphyosemion s.s. consists of two major clades. Aphyosemion musafirii is
in a clade with A. castaneum, A. polli, A. lamberti, A. rectogoense, and A. congicum. The distribution of all species of
Aphyosemion s.s. is discussed.
Resumé. Aphyosemion musafirii, nouvelle espèce, est décrite à partir de spécimens récoltés près de Ubundu (rivière Rui-
ki, bassin du Congo, Province Tshopo, République Démocratique du Congo). Une autre population de A. musafirii est
connue de la rivière Romée, 50 km à l’Ouest de Kisangani. Les rivières Ruiki et Romée sont de petits affluents rive gauche
du fleuve Congo. Aphyosemion musafirii peut être distingué de son plus proche parent A. castaneum par le patron de col-
oration mâle. Une analyse ADN préliminaire démontre que Aphyosemion s.s. consiste en 2 clades majeurs. Aphyosemion
musafirii est dans un clade avec A. castaneum, A. polli, A. lamberti, A. rectogoense et A. congicum. La distribution de
toutes les espèces de Aphyosemion s.s. est discutée.
Key words. Killifish, eastern Congo basin, Ubundu, systematics, taxonomy, biogeography.
Bonn zoological Bulletin Volume 60 Issue 1 pp. 73–87 Bonn, May 2011
and may well represent A. schoutedeni. Aphyosemion
christyi is restricted to the Epulu and Ituri drainages (about
350 km east of Kisangani) and is the only species that can
be identified by morphological characters (i.e., higher
number of dorsal fin rays than all other species)
(Boulenger, 1915; Van der Zee & Sonnenberg, unpubl. da-
Aphyosemion castaneum has long been misidentified as
A. christyi in literature and only recently Van der Zee &
Huber (2006) demonstrated that A. christyi is restricted to
elevations above 500 m over a distance of 180 km north-
east of the type locality, Bafwasende. Aphyosemion cas-
taneum is widespread around Kisangani, but additional
records of this species are known from Salonga Park by
Scheel (1990), who didn’t specify the exact location with-
in the park, Lompole, and Yaka in the central Congo Basin
(Huber, 2005b). According to Huber (2005b) it is likely
that these populations represent a different, undescribed
species. Consequently, A. castaneum might be restricted
to the right bank of the Congo River (Huber, 2005b), with
the exception of some populations that were found close
to the left bank, and the specimens collected on the left
bank with a wide distribution in the central basin were dis-
tinguished here as A. sp. aff. castaneum (Fig. 1).
From the Romée and Ruiki Rivers on the left bank of the
Congo River in the eastern part of the central Congo
drainage, three collections of Aphyosemion are known
from the MRAC, originally identified as A. christyi. The
colour pattern of the anal and caudal fins in preserved
specimens differs from A. castaneum, which is widespread
around Kisangani. In 2007, A. Van Deun (Leuven, Bel-
gium) collected a species of Aphyosemion with the same
colour pattern as the three previous collections from the
Romée and Ruiki Rivers, at two localities just north of
74 Jouke R. Van der Zee & Rainer Sonnenberg
Bonn zoological Bulletin 60 (1): 73–87 ©ZFMK
Fig. 1. Map showing the distribution of the genus Aphyosemion. Type localities are indicated by a T in the species symbol. The
type locality for A. congicum is unknown and for A. lefiniense and A. plagitaenium only collections from the type locality or in
proximity are currently known.
Ubundu on the left bank of the Congo River. Based on
live male colour pattern and a preliminary mitochondrial
DNA analysis this species is described here as
Aphyosemion musafirii, new species.
Morphometric measurements were taken with a digital cal-
liper, partly under a dissecting microscope, and rounded
to the nearest 0.1 mm. Counts and methods follow Ami-
et (1987). Measurements, including subunits of head, are
presented as percentages of standard length (SL). The
number of all visible rays of dorsal, anal, caudal, pelvic,
and pectoral fins were counted, the abbreviation D/A
means the relative position of the first dorsal fin ray with
regard to the opposite anal fin ray. Count of scales on the
mid-longitudinal series is the number of scales between
the upper attachment of the opercular membrane and the
caudal fin base. Excluded are the scales posterior to the
hypural junction, which were counted separately. Nomen-
clature for the neuromast system on the head follows
Scheel (1968) and Van Bergeijk & Alexander (1962), and
that for the supraorbital (frontal) squamation follows
Hoedeman (1958).
Total DNA was extracted from fin clips or muscle tissue
from the caudal peduncle of ethanol preserved specimens,
following a modified DNA extraction protocol after Gus-
tinicich et al. (1991). Specimens used for DNA analyses
are listed in Table 1 with GenBank accession numbers. A
fragment of the mitochondrial cytochrome b gene was se-
quenced for 25 specimens of Aphyosemion and Meso-
aphyosemion cameronense (GenBank accession number
A new Aphyosemion from the eastern Congo
Bonn zoological Bulletin 60 (1): 73–87 ©ZFMK
Table 1. List of specimens used for the DNA analyses with locality information and GenBank accession numbers. Abbreviations:
DRC = Democratic Republic of Congo; RCA = Republic of Central Africa; US = sample provided by Uli Schliewen, ZSM, Mu-
nich; AS = aquarium bred strain; CI = commercial import; WC = wild caught sample.
Species sample no. Country Collection locality GenBank acc. no.
Aphyosemion castaneum AS RS1408 DRC Kisangani JF307802
A. castaneum AS RS1499 Republic Congo Oyo JF307797
A. castaneum WC RS1790 DRC AVD 3 JF307803
A. cf. chauchei AS RS1527 Republic Congo Olombo JF307796
A. cf. decorsei AS RS1521 RepublicCongo Lobaye JF307795
A. cognatum WC RS1515 Republic Congo Mbonza II (US 107) JF307791
A. cognatum AS RS1520 DRC Lake Fwa JF307793
A. cognatum AS RS1529 DRC Kinsuka JF307794
A. congicum AS RS1617 DRC Z 82/17 JF307798
A. elegans WC RS1747 DRC Boende CI 2006 JF307792
A. elegans WC RS1513 DRC Inongo (US 24) JF307790
A. elegans WC RS1514 DRC Inongo (US 66) JF307789
A. lamberti AS RS1256 Gabon BSWG 97/9 JF307781
A. musafirii WC RS1787 DRC AVD 1 JF307804
A. polli AS RS1584 DRC CI JF307800
A. polli AS RS1479 RCA RCA 91/1, Kapou 1 JF307801
A. rectogoense AS RS1419 Gabon PEG 95/16 JF307799
A. sp. aff. castaneum WC RS1506 DRC Lompolé (US 74) JF307782
A. sp. aff. castaneum WC RS1507 DRC Lompolé (US 79) JF307783
A. sp. aff. castaneum WC RS1510 DRC Yaka (US 33) JF307786
A. sp. aff. castaneum WC RS1511 DRC Yaka (US 45) JF307787
A. sp. aff. castaneum WC RS1512 DRC Yaka (US 61) JF307788
A. sp. aff. elegans WC RS1508 DRC Lui Kotalé (US 75) JF307784
A. sp. aff. elegans WC RS1509 DRC Lui Kotalé (US 77) JF307785
A. sp. Cuvette AS RS1019 DRC Boende 2002 JF307780
Mesoaphyosemion cameronense RS262 Cameroon CMM 40 AY748282
76 Jouke R. Van der Zee & Rainer Sonnenberg
Bonn zoological Bulletin 60 (1): 73–87 ©ZFMK
Table 2. Uncorrected p-distances for the mitochondrial cytochrome b sequences with pairwise exclusion of missing data. Abbreviations for the DNA samples: CAM = M. came-
ronense, CAS = A. castaneum, cCHA = A. cf. chauchei, COG = A. cognatum, CON = A. congicum, cDEC = A. cf. decorsei, ELE = A. elegans, LAM = A. lamberti, MUS = A. mu-
safirii, POL = A. polli, REC = A. rectogoense, sCAS = A. sp. aff. castaneum, sELE = A. sp. aff. elegans, spCU = A. sp. Cuvette.
AY748282, published in Sonnenberg & Blum [2005]) was
used as outgroup, for lab protocols see Sonnenberg et al.
Resulting sequences were aligned with ClustalX 1.8
(Thompson et al., 1997) and checked by eye in BioEdit (Hall, 1999). All sequences were translated into the
corresponding amino acids and tested for the anti-G bias
of mitochondrial sequences (Zhang & Hewitt, 1996) to
confirm for functionality and mitochondrial origin. Un-
corrected p-distances with pairwise exclusion of missing
data were calculated in MEGA 4.1 beta 3 (Tamura et al.,
2007) and are given in Table 2.
Analyses of sequence data were performed with PAUP
4.0b10 (Swofford, 1998) by maximum parsimony and
with MrBayes 3.1.2 (Huelsenbeck & Ronquist, 2001) by
a Bayesian approach, all analyses with the identical respec-
tive settings given in Sonnenberg & Busch (2010). In ad-
dition, we performed a maximum parsimony analysis with
the same parameters as the previous, but with exclusion
of the third protein coding position. Only nodes, which
are supported with bootstrap values above 75% or poste-
rior probabilities above 95% were considered as support-
ed by the data.
We used the software SplitsTree (Huson & Bryant, 2006)
to calculate a split decomposition network representation
of the dataset.
As species concept we adopted the approach by Moritz
et al. (2000), which is similar to the Evolutionary or Phy-
logenetic species concepts (as discussed in Kottelat, 1997).
Aphyosemion musafirii, new species
(Figs 2–6, Tables 3–5)
Holotype. MRAC 2011-007-P-1, male, 36.7 mm SL,
Democratic Republic of Congo, Tshopo Province, 67 km
on the road from Kisangani to Ubundu (1°30’ N, 25°21’
E), 450 m altitude, north-eastern Congo basin, A. Van De-
un, 22 October 2007.
Paratypes. MRAC 2011-007-P-2-5, 4 females, 30.1–33.5
mm SL, collected with the holotype; MRAC 2011-007-
P-6-9, 4 males, 32.9–34.6 mm SL, collected with the holo-
Additional non-type material. MRAC 90-30-P-1471, la-
belled as A. christyi (Boulenger, 1915), Democratic Re-
public of Congo, Ubundu, L. De Vos, February 1990.
MRAC 90-47-P-846-853, labelled as A. christyi
(Boulenger, 1915), Democratic Republic of Congo, Riv.
Romée, km 30 route Kisangani-Opala, L. De Vos & C. Da-
nadu, 07.02.1990.
A new Aphyosemion from the eastern Congo
Bonn zoological Bulletin 60 (1): 73–87 ©ZFMK
Fig. 2. Aphyosemion musafirii, male, collected with the
types by A. Van Deun, 22.10. 2007, not preserved. Type local-
ity, 67 km on the road from Kisangani to Ubundu, Democratic
Republic of Congo. Photo: H. Ott.
Fig. 3. Aphyosemion musafirii, male,7 km north of Ubundu,
Democratic Republic of Congo, collected by A. Van Deun,
22.10.2007, not preserved. Photo: H. Ott.
Fig. 5. Aphyosemion musafirii, female, 7 km north of Ubun-
du, Democratic Republic of Congo, collected by A. Van Deun,
22.10.2007, not preserved. Photo: H. Ott.
Fig. 4. Aphyosemion musafirii, male, F1 from specimens col-
lected 7 km north of Ubundu, Democratic Republic of Congo,
showing a different anal fin pattern than most of the wild col-
lected males from that population. Photo: W. Grell.
MRAC 90-47-P-854-863, labelled as A. christyi
(Boulenger, 1915), Democratic Republic of Congo, Riv.
Romée km 3, rive gauche, route Kisangani-Opala, L. De
Vos & C. Danadu, 10.–17.04.1990.
Diagnosis. Aphyosemion musafirii (Fig. 2–6) is placed
within the genus Aphyosemion by the combination of the
following characters: preopercular neuromast system with
6 pores, slender body, posterior origin of dorsal fin with
less than 10 fin rays, females with a strong reticulation due
to dark scale borders, and the extended edges of the cau-
dal fin in males (Huber, 2005a).
Males of A. musafirii differ from all other representatives
of Aphyosemion in the north-eastern Congo Basin by the
in average broader red margin of the dorsal fin. This red
border is narrow (up to 10%) in all other species in this
area and is often absent at the distal end of the fin.
Males of A. musafirii differ from A. castaneum by the ab-
sence of a red band in the approximate centre of the anal
fin, absence of a red infra-buccal band, absence of red
edges around the light ventral zone in the caudal fin, ab-
sence of, or only very narrow, yellow margin of pelvic fins
and a higher average number of red dots on side (A.
musafirii: min. 51, max. 117, average = 79, sd = 23, num-
ber of specimens = 11; A. castaneum: min: 13, max. 57,
average = 30, sd = 15, number of specimens = 20). These
dots are arranged in more or less regular, interrupted rows
in A. musafirii, whereas these dots are irregularly distrib-
uted in A. castaneum. The higher number of red dots on
the side also distinguishes males of A. congicum (25–40),
A. schoutedeni (17–28) and A. polli (11–49) from A.
musafirii. It can also be distinguished by the more or less
regular interrupted rows of red dots from A. sp. Cuvette,
A. elegans (Boulenger 1899), and A. plagitaenium, which
have either a pattern of vertical streaks or oblique bars on
the side in males. Males of A. musafirii differ from A.
schoutedeni by the same characters as A. castaneum, with
the exception that A. schoutedeni lacks a red band in the
anal fin. Males of A. musafirii have less spots in the cau-
dal fin than A. christyi (35–55 in A. musafirii; over 70 in
A. christyi). The background colour of the flanks is blue-
green in A. musafirii and purple-blue in A. christyi. In ad-
dition to colouration characters, males of A. musafirii can
be distinguished from A. christyi by the lower number of
dorsal fin rays (7–9 in A. musafirii versus 10–11 in A.
Description. See Figures 2–6 for general appearance and
Tables 3–5 for morphometric and meristic data of the type
series. Aphyosemion musafirii shows strong sexual dimor-
phism, males more colourful, unpaired fins larger, dorsal
and anal fins with posterior fin rays extended. A medium
sized, slightly laterally compressed species; dorsal profile
slightly convex, greatest body depth approximately at
pelvic fins. Ventral profile slightly convex from head to
end of anal fin, concave on caudal peduncle. Snout slight-
ly rounded, mouth directed upwards, lower jaw longer
than upper jaw. Dentary bears an outer row of large and
inner irregular rows of smaller unicuspid, curved teeth; the
premaxilla bears some larger and several smaller unicus-
pid and curved teeth.
Frontal (after Scheel, 1968) or nasal (after van Bergeijk
& Alexander, 1962) neuromasts in separate grooves, the
preopercular canal with six pores.
Scales cycloid, entirely scaled except ventral surface of
head; frontal squamation of G-type; scales on mid-longi-
tudinal series 29–30, with 1–2 scales posterior to the hy-
pural plate; 7 transversal scales, 12 scales around the cau-
dal peduncle.
78 Jouke R. Van der Zee & Rainer Sonnenberg
Bonn zoological Bulletin 60 (1): 73–87 ©ZFMK
Fig. 6. Variability of the anal fin pattern in Aphyosemion
musafirii (all examples from wild caught specimens).
Small dorsal fin with 7–9 fin rays, first dorsal fin ray in-
serts above the 7–10th anal fin ray; anal fin with 13–15
rays; posterior dorsal and anal fin rays slightly elongated
in males; caudal fin with 21–24 rays, with extensions on
upper and lower fin rays. Pectoral fin with 12–14, pelvic
fin with 5 rays.
Live colouration Males. (Figs 2–4 & 6) Flanks greyish
brown with blue-green iridescence. Edges of scales on the
flanks with dark pigmentation resulting in a reticulated
pattern. Dorsally the scales have broader pigmented edges
than ventrally. Flanks with approximately 50 to 120 red
spots. These spots are mainly situated at the anterior edge
of the scales and are mostly irregularly distributed in lines,
usually forming up to five parallel lines. Three red streaks
on opercle in an approximate 45° angle. Infra-buccal band
absent or only present at the sides of the jaw.
Pectoral fin yellow, unspotted or just with some tiny spots
in the centre. Pelvic fin yellow with several prominent red
spots. Dorsal and anal fin blue iridescent at the base and
yellow distally, provided with red spots; spots more nu-
merous and larger at the base than distally. Dorsal fin
edged with a broad dark red band (up to 30% of fin length
at mid section). Anal fin edged with a narrow red band.
Spots on anal fin rounded to elongated, sometimes miss-
ing at the distal part of the fin, leaving a broad yellow sub-
distal band. Caudal fin light blue with yellowish distal
edges provided with rounded and/or 35–55 elongated red
spots. Dorsal and ventral edge of the caudal fin provided
with a broad dark red band.
Females. (Fig. 5) Flanks grey with darker reticulation. Red
streaks on opercle reduced and infra-buccal band absent.
All fins transparent. Edge of anal fin and distal part of ven-
tral fins light blue. Faint spots on anal fin, more promi-
nent spots on base of dorsal and dorsal part of caudal fin.
After one year of preservation in ethanol. Males. Flanks
light brown with transition to light ventral side. Reticu-
lation, buccal band, and red streaks on opercle as in live
specimens. Spots on flanks as in live specimens, but small-
er and more vague. All fins transparent greyish, provid-
ed with spots as in live specimens, but vague and pink.
Females. As in live specimens, but spots on anal fin hard-
ly visible.
Distribution. (Fig. 1) Aphyosemion musafirii is restrict-
ed to brooks in the Ruiki and Romée River systems on
the left bank of the Congo River between Kisangani and
Ubundu, Tshopo Province, Democratic Republic of Con-
A new Aphyosemion from the eastern Congo
Bonn zoological Bulletin 60 (1): 73–87 ©ZFMK
Table 3. Morphometrics of Aphyosemion musafirii, new species, (H = holotype, P = paratypes: 4 males and 4 females ). All mea-
surements in percentages of standard length, standard length in mm.
H ♂ P ♂ P ♂ P ♂ P P ♀ P♀ P ♀ P ♀
Standard length 36.7 34.6 34.2 33.3 32.9 33.5 32.6 31.3 30.1
Body depth 21.0 18.4 19.6 19.5 18.8 18.8 20.8 20.1 21.4
Head length 21.5 20.5 21.6 21.0 20.4 19.1 20.5 19.5 19.0
Eye diameter 6.3 6.3 6.4 6.3 6.1 5.7 6.4 6.4 6.9
Interorbital width 12.0 12.7 11.1 11.4 12.5 10.7 11.3 11.2 13.0
Pre-dorsal length 63.2 61.0 66.0 66.4 62.1 60.2 66.2 65.8 63.5
Pre-anal length 55.0 54.9 53.5 54.3 50.8 55.2 56.1 54.6 57.3
Dorsal fin base 11.2 12.1 10.2 11.1 11.2 11.0 10.1 10.9 11.0
Anal fin base 19.1 18.8 17.5 19.2 18.2 17.3 16.1 17.3 17.2
Caudal peduncle depth 12.5 11.5 11.1 11.7 11.6 10.7 11.7 11.5 11.9
Table 4. Meristics of Aphyosemion musafirii, new species.
Numbers indicate observed values; numbers in parenthe-
ses frequency of occurrence; values found for the holo-
type are indicated by an asterisk.
meristic count values (frequency)
dorsal fin rays 9 (3*), 8 (4) 7 (2)
anal fin rays 13 (5*), 14 (3), 15(1)
D/A +7 (1*), +8 (1), +9 (3), +10 (4 fem.)
caudal fin rays 21 (3*), 22 (2), 23 (3), 24 (1)
pelvic fin rays 5 (9*)
pectoral fin rays 12 (2*), 13 (6), 14 (1)
lateral line scales 29 (2), 30 (6*), 31 (1)
transversal row scales 7 (9*)
circumpenduncul. scales 12 (9*)
Etymology. Aphyosemion musafirii is named after Dr.
Jean Musafiri (Ubundu, Democratic Republic of Congo),
coordinator for the national tuberculosis and leprosy con-
trol programme in the “Province Orientale Occidentale”,
the huge forest area around Kisangani. The name Musafiri
means “traveller” in Swahili. Indeed, he travels around the
area under very difficult circumstances, covering enor-
mous distances by jeep, motorised canoe or small motor-
cycle. Dr. Musafiri was born in Ubundu and has always
stayed in the province to help his people, in spite of the
very difficult living conditions and the atrocities of the war
in the eastern Democratic Republic of Congo. He made
it possible that the type material of A. musafirii and a new
species of Fenerbahce (Sonnenberg, Woeltjes & Van der
Zee, submitted) could be collected by A. Van Deun (In-
stitute of Tropical Medicine, Antwerp) at the occasion of
an external evaluation visit of the tuberculosis/leprosy pro-
DNA analyses. The resulting sequence alignment has a
final length of 760 bp, the base composition shows the,
for mitochondrial sequences typical, A/T bias (Zhang &
Hewitt, 1996). In two sequences (RS1747 and RS1521),
up to 13 N were introduced at the start of the alignment
for equal sequence length. We found 192 variable and 130
phylogenetic informative positions. The DNA fragment
translates into 253 amino acids, of which 27 are variable
and 15 phylogenetic informative, and contains no unex-
pected stop codon. Uncorrected pairwise distances be-
80 Jouke R. Van der Zee & Rainer Sonnenberg
Bonn zoological Bulletin 60 (1): 73–87 ©ZFMK
Table 5. Comparison of the morphometric values for Aphyosemion musafirii, new species, and members of the sub-
genus Aphyosemion. Abbreviations used in the table: CAS = A. castaneum, CHR = A. christyi, COG = A. cognatum,
CON = A. congicum, LEF = A. lefiniense, MEL = A. melanopteron, MUS = A. musafirii, POL = A. polli, TEU = A.
teugelsi, E = eye diameter, I = interorbital width, BD = body depth, HL = head length, HW = head width, pD = pre-
dorsal fin distance, pA = preanal fin distance, DB = dorsal fin base, AB = anal fin base, CD = caudal peduncle depth,
sd = standard deviation.
species location sex SL E I BD HL pD pA DB AB CD
COG Z 91/3, 33.5 8.0 11.0 22.9 18.9 67.1 62.2 11.5 21.9 14.3
CHR HZ 85/14, 29.1 7.9 9.3 18.4 21.9 65.7 59.0 11.6 19.6 11.2
Epulu 31.2 7.8 9.3 17.1 22.1 64.5 58.7 11.3 19.9 10.9
POL RCA 91/1, 36.0 7.5 10.6 22.5 23.3 68.3 59.4 12.1 21.7 13.1
Kapou 29.1 6.9 8.2 23.6 22.7 67.0 61.9 11.7 19.9 12.6
CON type of MEL 7.8 9.5 21.6 27.8 67.3 61.5 12.1 20.9 13.4
CAS HZ 85/13 20.8 7.7 13.4 20.4 20.2 70.2 53.8 12.0 22.1 11.1
19.4 7.7 11.9 20.8 19.8 68.3 52.1 11.5 20.0 12.3
LEF Luna River 23.1 8.7 12.1 19.0 24.1 66.6 54.9 7.0 13.0 9.6
19.2 8.3 11.4 21.0 23.6 69.9 57.0 8.6 16.2 11.8
mean 7.8 10.6 20.8 22.2 67.7 58.1 11.0 19.6 12.2
sd 0.5 1.4 2.0 2.6 1.9 3.4 1.6 2.7 1.4
range 6.9–8.7 8.2–13.4 17.1–23.6 18.9–27.8 64.5–70.2 52.1–62.2 8.6–12.1 13.0–22.1 9.6–14.3
mean 6.3 12.1 19.2 30.9 68.3 61.1 14.6 19.0 12.4
sd 0.3 0.5 2.2 1.0 1.6 1.5 0.7 1.2 0.2
range 5.4–16.9 11.3–12.8 16.6–21.2 29.3–32.0 66.0–70.5 59.2–62.8 13.8–15.3 17.4–20.7 12.1–12.7
Fig. 7. Aphyosemion castaneum, 11 km west of Kisangani,
Democratic Republic of Congo, not preserved. Photo: H. Ott.
Fig. 8. Aphyosemion castaneum, Wani Rukula, 65 km south-
east of Kisangani, Democratic Republic of Congo. Photo: H. Ott.
A new Aphyosemion from the eastern Congo
Bonn zoological Bulletin 60 (1): 73–87 ©ZFMK
Fig. 9. Phylogenetic tree by maximum parsimony, tree length = 283, consistency index = 0.5194, homoplasy index = 0.4806,
retention index = 0.7467, and rescaled consistency index = 0.3879. Shown is one of 36 equally short trees, on the left side of nodes
support values are shown for maximum parsimony bootstrap analyses and Bayesian posterior probabilities in the following order:
maximum parsimony / maximum parsimony with exclusion of third position / Bayes with Nst = 2 / Bayes with Nst = 6. Only sup-
port values for nodes are shown, for which at least in one analysis a bootstrap value of 75% or posterior probabilities of 95% were
reached. No value given means that the support value in all analyses is below the previous given values or the node is not recov-
ered in the analysis. Some values are given below the tree and are indicated on the respective nodes by a capital letter for better
tween sequences are presented in Table 2, the observed
values found between Aphyosemion and Mesoa-
phyosemion as the outgroup range from 12.6–14.9%, the
maximum observed value within Aphyosemion is 9.1%.
The samples of A. cognatum from aquarium strains orig-
inating from Kinsuka and Lake Fwa are, despite nearly
900 km distance, identical, and also with the shorter se-
quence of an A. cognatum from Lake Fwa published by
Murphy & Collier (1999, GenBank acc. no. AF002324),
therefore it might be possible that the strains were erro-
neously mixed or mislabelled since their introduction in
the killifish hobby. A test with a second specimen from
the Kinsuka strain resulted in the same sequence (data not
The A. cognatum 1515 sample is only represented by a
female, so the species identification is tentative based on
the knowledge of other Aphyosemion collections around
this area and the resulting sequence is identical with a
shorter sequence of Murphy & Collier (1999, GenBank
acc. no. AF002327), which they have published as A. el-
egans from Naoimda, and slightly different from their se-
quence of what they call A. elegans from Madimba (Gen-
Bank acc. no. AF002328). It is not clear if Naoimda is a
misspelling of Madimba. However, the specimens from
Madimba belong to A. cognatum (e.g. see Seegers, 1997,
p. 74), the sequences differ in only two bases.
The phylogenetic analyses (Fig. 9) give an only partially
resolved tree of the analysed specimens. In addition, there
are differences in support for nodes; excluding the third
protein coding positions gives no bootstrap support above
50% for several nodes, which were recovered by the oth-
er analyses (Fig. 9). Aphyosemion musafirii seems to be
closer to A. castaneum, however, this node is only sup-
ported by the maximum parsimony analyses. Comparing
the phylogenetic hypotheses summarized in the tree (Fig.
9) with the split-decomposition network representation
(Fig. 10), the network indicates a clear separation of the
analysed specimens into two main groups, which is not
that prominent in the dichotomous phylogenetic hypothe-
ses (Fig. 9). One group, in the following called the A. cas-
taneum group (CAS) is characterised by longer branches
between the species (range 1.1–8.2% p-distance), where-
as the second group, the A. cognatum group (COG) has
82 Jouke R. Van der Zee & Rainer Sonnenberg
Bonn zoological Bulletin 60 (1): 73–87 ©ZFMK
Fig. 10. SplitsTree network representation of the Aphyosemion dataset. Number of taxa = 26, 760 base pairs, fit = 98.01, uncor-
rected p-distances, neighborNet.
shorter internal branch length (0.0–4.1% p-distance), the
divergence between both groups ranges between 5.8–9.1%
p-distance. The CAS group contains, according to the net-
work (Fig. 10), five species, which were found from the
western to the eastern Congo Basin, and two species from
Gabon. However, this group is not recovered by the phy-
logenetic analyses with any significant support (Fig. 9).
The second group in the network (Fig. 10) is recovered
with high support (Fig. 9) at least by three analyses and
contains samples from the western and central Congo
Colour pattern
The main criterion for distinguishing species in notho-
branchiids and especially Aphyosemion is the male
colour pattern. All Nothobranchiidae show a polygamous
mating system and a high degree of sexual dimorphism.
Amiet (1987) therefore assumed that male colouration
might be important in mate choice of females in notho-
branchiid fishes. Van der Zee et al. (2007) assumed that
especially the colour pattern of the caudal peduncle and
unpaired fins plays an important role in female mating
preferences. Kullmann & Klemme (2007) were able to
demonstrate for Chromaphyosemion that females prefer
to mate with their own males, for Diapteron this was stud-
ied by Brosset & Lachaise (1995). Thus female mate
choice on divergent male colour characters may enable
speciation by sexual selection. After secondary contact of
previously allopatric populations, species cohesion is then
easily maintained in parapatry or sympatry.
The current study shows that the genus Aphyosemion re-
mains taxonomically problematic, which is in part due to
old type material with no traces of the most important
characters, the male colouration pattern, and descriptions,
which do not give detailed information about live or pre-
served colouration. On the other hand, many species show
a certain degree of variation in colour pattern within and
between populations, which makes identification in some
cases difficult. Differing species identification between au-
thors further complicates killifish literature, as for exam-
ple in the case of the A. elegans samples in the study of
Murphy & Collier (1999) mentioned above. A third sam-
ple listed as A. elegans in their study, an aquarium strain
from Epoma (Murphy & Collier, 1999) was later described
as A. plagitaenium by Huber (2004).
Our preliminary molecular phylogeny of this group gives
a very complex pattern. The CAS (A. castaneum) group
might be not a monophyletic unit, but consists of sever-
al species with deep divergences, which is reflected by the
large p-distances found in this group (Table 2). The root
of the group might be placed close to this species com-
plex (Fig. 10). The COG (A. cognatum) group consists of
species with lower within-group p-distances (Table 2), ap-
pears to be monophyletic and might contain species of
more recent origin. Both groups cover large ranges in the
Congo and adjacent river basins in Gabon (A. lamberti and
A. rectogoense), have a kind of mixed distribution and oc-
cur in some cases in sympatry (e.g. A. elegans and A. sp.
Cuvette) or even syntopically (A. sp. Cuvette and A. sp.
aff. castaneum; A. chauchei and A. elegans; A. “schioet-
zi” and A. polli).
Interestingly some phenotypically different species (A. sp.
aff. castaneum and A. elegans / A. sp. aff. elegans) in the
Cuvette centrale turn out to be closely related and not ge-
netically separated by the mitochondrial DNA data (Fig.
9). Compared with some additional nuclear 28S rDNA
(LSU) sequences (unpublished data), it turns out that the
studied samples can be separated into two groups: one in-
cluding A. castaneum and A. sp. aff. castaneum and on
the other hand the samples of A. elegans / A. sp. aff. ele-
gans. This indicates that most probably mitochondrial in-
trogression between species of the CAS and COG group
has occurred in an area were both groups are at least para-
patric. However, despite the potential of hybridization as
is indicated by the mitochondrial introgression, both
species or species groups live at least in sympatry and have
stable distinguishing characters.
Distribution of the genus Aphyosemion
With the addition of A. musafirii the number of species
in the eastern Congo Basin is increased to four (see In-
troduction). The study of the Aphyosemion specimens in
the MRAC by the first author also added some collections
of other species from this area (Fig. 1), in part due to
misidentified samples. However, the maximum number of
eight species is found in the westernmost part of the Con-
go Basin. An artificial factor, which might explain the con-
centration of species here, is the higher number of sam-
ples from this area. A natural cause of high species diver-
sity might be the proposed existence of a forest refuge dur-
ing dryer periods in the past (Leal, 2004; Maley, 1996;
Plana, 2004). Two of these eight species are widely dis-
tributed in the northern part of the basin: A. castaneum
Myers, 1924 and A. polli Radda & Pürzl, 1987. These two
species were probably transported downstream by the
Congo and Ubangui rivers, and both can also be found at
Pool Malebo. Aphyosemion elegans is here present in the
northern tributaries of the Likouala River, but is wide-
spread in the central Congo Basin (Cuvette centrale) and
north of the middle Congo section up to Buta.
A new Aphyosemion from the eastern Congo
Bonn zoological Bulletin 60 (1): 73–87 ©ZFMK
Three species have a more restricted distribution in the
Central Republic of Congo: A. chauchei Huber & Scheel,
1981 is found in the Likouala and Alima drainages, A.
lefiniense Woeltjes, 1984 is restricted to the southern trib-
utaries of the Lefini River, and A. plagitaenium Huber,
2004 is only known from the type locality at Epoma in
the northern Central Republic of Congo.
Aphyosemion schioetzi Huber & Scheel, 1981 has a dis-
junctive distribution: a rather restricted area in the south
around the type locality and a huge area in the north, sep-
arated by a gap of almost 280 km where only A. lefiniense
is found (Huber & Scheel, 1981). The present authors sug-
gest that the northern populations (labelled here in the fol-
lowing “schioetzi”, Fig. 1) do not belong to A. schioetzi
sensu strictu, since all representatives are lacking the dark
red edge of the anal fin and seem to be built more slen-
der than A. schioetzi. The northern populations might rep-
resent a new species or be conspecific with A. decorsei
(Pellegrin, 1904). The status of A. decorsei has been sub-
ject to discussion for a long time (Huber, 1994, 2004,
2005a; Scheel, 1968, 1990; Wildekamp, 1993). Poll (1951)
even placed it in Epiplatys and in the description of Hap-
lochilus decorsei Pellegrin assumed it to be close to
Aplocheilichthys spilauchen. Myers (1924) placed it in
Aphyosemion with some hesitation. Scheel (1968, 1990),
Huber (1994, 2004, 2005a), and Wildekamp (1993) all
have seen the types and they confirmed Myers’s statement.
The types from the Central African Republic are in poor
condition with the colour pattern lost. Huber (2004, 2005a)
stated that A. decorsei has only few red spots on the flanks
and might be conspecific with A. polli, the latter being a
junior synonym. Wildekamp (1993), however, is con-
vinced that A. decorsei has many spots on the flanks, based
on the light spots on the scales of the syntypes. Red pig-
mentation pattern, after preservation in formalin and trans-
fer in ethanol, leaves corresponding patterns of lighter ar-
eas than the body colouration (Van der Zee & Sonnenberg,
2010). Aphyosemion polli has not only few spots on the
flanks, but also has very few spots (or no spots) on the
anal fin, arranged near the base of this fin. In the original
description of A. decorsei Pellegrin writes: “la dorsale,
l’anale et des ventrales avec des petits points carmins plus
ou moins nombreux” (dorsal, anal, and ventral fins with
small more or less numerous carmine spots). The present
authors agree with Wildekamp’s argumentation: A. decor-
sei is a species with many spots at least on the anal fin.
Aphyosemion cognatum Meinken, 1951 has a huge dis-
tribution area in the south from the right bank of the Con-
go River to Lodja in the upper Lukenie basin. Sympatric
with A. cognatum occur A. ferranti (Boulenger, 1910), A.
congicum (Ahl, 1924), and A. lujae (Boulenger, 1911).
In the large Cuvette centrale only three species occur: a
blue species that resembles A. castaneum (as A. sp. aff.
castaneum in Fig. 1), A. elegans, and another undescribed
species (A. sp. Cuvette) that can occur sympatrically with
A. elegans. This species, with a characteristic dark dorsal
fin, has long been taken for A. elegans in many publica-
tions (Radda & Pürzl, 1987; Huber, 2004, 2005a,b;
Wildekamp, 1993). However, the description and origi-
nal drawing do not mention this dark dorsal fin that is very
prominent even in preserved specimens (Boulenger,
1899). Our DNA analysis shows that A. elegans and A.
sp. Cuvette are not closely related (Fig. 9). Additionally,
the specimens from Lui Kotale (Fig. 1) differ by male
colour pattern and are tentatively named A. sp. aff. ele-
gans to indicate the differences and their potential status
as separate species.
Historical influcences on current pattern
The distribution pattern within the group is very complex.
Several species are found over large areas (Fig. 1), often
mixed with congeners. This complexity is probably relat-
ed to the history of the Congo Basin.
From our data it cannot be determined when Aphyosemion
reached the Congo Basin. The ancestors of the extant
species must have come from the west where their clos-
est relatives live. With the current data it is not possible
to decide if they were already dispersed around the en-
dorheic lake or if they entered the basin after a river cap-
ture to the west. Both scenarios will result in different dis-
tribution and dispersal patterns.
In later stages forest refuges probably played an impor-
tant role in the establishment of the mosaic distribution
pattern of Aphyosemion during glacial dry periods. Almost
all species of Aphyosemion s.l. are strictly bound to for-
est cover (Brosset, 1982; Kamdem Toham & Teugels,
1997, 1998, 1999). During glacial periods, the majority
of the forest in the Congo Basin was replaced by savan-
nah (see Leal, 2004; Maley, 1996). Only in the western
part of the basin a relatively large refuge was present from
where dispersal might have originated during more humid
periods. Leal (2004) proposed that besides larger well
known refuges also micro refuges must have existed, e.g.
in Gabon. A combination of one or more larger and sev-
eral micro refuges from where repeatedly dispersal could
have occurred might explain the very complex distribu-
tion pattern of Aphyosemion in the Congo Basin.
Further DNA studies can contribute to alpha taxonomy and
towards a better understanding of the phylogeny and bio-
84 Jouke R. Van der Zee & Rainer Sonnenberg
Bonn zoological Bulletin 60 (1): 73–87 ©ZFMK
geography of this group and add to our understanding of
its evolution. However, much more samples from the huge
distribution area of Aphyosemion will be needed and the
occurrence of mitochondrial introgression makes it nec-
essary to include several nuclear markers into such a study
to get reliable and well supported results.
Part of the comparative material is listed in Van der Zee
& Sonnenberg (2010); additional material is listed here
and in the Online Appendix:
Aphyosemion sp. Cuvette: MRAC 79-09-P-720-722, la-
belled as A. christyi (Boulenger, 1915), Democratic Re-
public of Congo, Iteli River, terr. Opala, J. Lambert,
Aphyosemion sp.: MRAC 90-30-P-1471, labelled as A.
christyi (Boulenger, 1915), Democratic Republic of Con-
go, Yangambi, J. Lambert, 08.05.1957; MRAC 119855-
856, labelled as A. christyi (Boulenger, 1915), Democrat-
ic Republic of Congo, Yangambi, J. Lambert, 09.05.1957.
Aphyosemion castaneum Myers, 1924: MRAC 22555-
22561, Democratic Republic of Congo, Stanleyville, Dr.
Richard, 1930; MRAC 89-043-P-547-612, labelled as A.
christyi (Boulenger, 1915), Democratic Republic of Con-
go, Libuku River near Kisangani, L. De Vos & M. Katem-
bo, April 1988.
Aphyosemion schoutedeni (Boulenger, 1920):
MRAC15664-15665, identified by David (1936) as Epi-
platys boulengeri, Democratic Republic of Congo, Med-
je, Lang & Chapin, 1910; MRAC 25529, Democratic Re-
public of Congo, Medje, H. Schouteden, no year.
Acknowledgements. The authors wish to thank the following
persons for their help in realizing this paper: J. Musafiri (Ubun-
du, Democratic Republic of Congo) for enabling the collection
of the new species; A. Van Deun (Leuven, Belgium) for the col-
lection of type material and several additional specimens that
were taken to Belgium alive and donated to H. Ott
(Mönchengladbach, Germany), who photographed them; W.
Grell (Neustadt-Erlach, Germany) for the picture of the F1 off-
spring of A. musafirii; W. Eberl (Haubersbronn, Germany) for
providing notes about a collecting trip in the northern Demo-
cratic Republic of Congo by H. Bleher and W. Grell in 1986; E.
Vreven and J. Snoeks (MRAC, Tervuren, Belgium) for their kind
hospitality and support; M. Parrent (MRAC) for his services and
ad hoc help during research at the museum; T. Woeltjes (Nij-
megen, The Netherlands) for access to his private library and
for providing data of hobbyist collection trips; U. Schliewen
(ZSM, Munich, Germany), R. Numrich (Cologne, Germany),
and the members of the Aphyosemion elegans AG of the Ger-
man Killifish Association (DKG), for DNA samples and valu-
able information. We also thank two anonymous reviewers, F.
Herder (ZFMK, Bonn, Germany), A. C. Schunke (MPI, Plön,
Germany), and B. Sinclair (Canadian National Collection of In-
sects, Ottawa, Canada) for comments and corrections on previ-
ous versions of the manuscript.
The Online Appendix is available at
Ahl E (1924) Über neue afrikanische Zahnkarpfen der Gattung
Panchax. Zoologischer Anzeiger 60: 303–312
Amiet JL (1987) Le Genre Aphyosemion Myers (Pisces, Tele-
ostei, Cyprinodontiformes). Faune du Cameroun, Volume 2.
Sciences Nat, Compiègne, France
Boulenger GA (1899) Matérieaux pour la fauna du Congo. Pois-
sons nouveaux du Congo. Cinquième Partie. Cyprins, Silures,
Cyprinodontes, Acanthoptérygiens. Annales du Musee du
Congo (Série Zoologie) Volume 1 (fasc. 5): 97–128, pls. 40–47
Boulenger GA (1910) Poisson Cyprinodonte nouveau du Bas-
Congo. Monatsberichte der Gesellschaft der Luxemburger Na-
turfreunde (Bulletins Mensuels de Société des Naturalistes
luxembourgeois) 4: 285
Boulenger GA (1911) Liste des poissons réprésentés dans une
nouvelle collection de la riviére Sankuru (Kasai), reçue de M.
E. Luja. Monatsberichte der Gesellschaft der Luxemburger
Naturfreunde (Bulletins Mensuels de Société des Naturalis-
tes luxembourgeois) 5: 223–224
Boulenger GA (1915) Catalogue of the freshwater fishes of
Africa in the British Museum (Natural History) Volume 3,
Boulenger GA (1920) Poissons recueillis au Congo belge par
l’expedition du Dr. C. Christy. Annales du Musée du Congo
belge: 1–38, pls 23–25
Brosset A (1982) Le peuplement de cyprinodontes du bassin de
l’Ivindo, Gabon. Terre Vie 36: 233–292
Brosset A, Lachaise D (1995) Evolution as a lottery conflicting
with evolution via sexual selection in African rainforest-
dwelling killifishes (Cyprinodontidae, Rivulinae, Diapteron).
Evolutionary Biology 28: 217–264
Collier GE (2007) The genus Aphyosemion: Taxonomic histo-
ry and molecular phylogeny. Journal of the American Killi-
fish Association 39: 147–168
Fowler HW (1936) Zoological Results of the George Vander-
bilt African Expedition of 1934. Part 3. The Freshwater Fish-
es. Proceedings of the Academy of Natural Science Philadel-
phia 88: 243–335, figs 102–152
Gustinicich S, Manfioletti G, Del Dal G, Schneider C, Caninci
P (1991) A fast method for high quality genomic DNA extrac-
tion from whole human blood. BioTechniques 11: 298–302
Hall TA (1999) BioEdit: a user-friendly biological sequence
alignment editor and analysis program for Windows 95/98/NT.
Nucleic Acids Symposium Series 41: 95–98
Hoedeman J (1958) The frontal scalation pattern in some groups
of toothcarps (Pisces: Cyprinodontiformes). Bulletin of
Aquatic Biology 1: 23–28
Huber J-H (1994) Aphyosemion passaroi, espèce inédite du
Gabon sud-oriental, au patron de coloration unique et descrip-
tion complémentaire de Aphyosemion decorsei (Pellegrin).
Cyprinodontiformes, Aplocheilidae. Revue française d’Aquar-
iologie 20: 77–80
A new Aphyosemion from the eastern Congo
Bonn zoological Bulletin 60 (1): 73–87 ©ZFMK
Huber J-H (2004) Description of a new Aphyosemion species
from Congo, A. plagitaenium n. sp., exhibiting a probable in-
tra-generic color convergence with oblique bars. Freshwater
and Marine Aquarium 27: 70–74
Huber J-H (2005a) Identifikation einer kleinen Sammlung von
Aphyosemion aus Zaire im Münchener Museum (Z.S.M.), mit
weiteren Kommentaren über die Validität der bekannten Kom-
ponenten der elegans-Superspezies. 1. Teil. Deutsche Killi-
fisch Gemeinschaft Journal 37: 8–21
Huber J-H (2005b) Identifikation einer kleinen Sammlung von
Aphyosemion aus Zaire im Münchener Museum (Z.S.M.), mit
weiteren Kommentaren über die Validität der bekannten Kom-
ponenten der elegans-Superspezies. 2. Teil. Deutsche Killi-
fisch Gemeinschaft Journal 37: 35–45
Huber J-H (2007) Killi-Data 2007. Updated checklist of taxo-
nomic names, collecting localities and bibliographic references
of oviparous cyprinodont fishes (Atherinomorpha, Pisces). So-
ciéte française d’Ichthyologie, Paris (German edition)
Huber J-H, Scheel JJ (1981) Revue systématique de la super-
espèce Aphyosemion elegans. Description de A. chauchei n.
sp. et A. schioetzi n. sp. Revue française d’Aquariologie 8:
Huelsenbeck JP, Ronquist F (2001) MRBAYES: Bayesian in-
ference of phylogeny. Bioinformatics 17: 754–755
Huson DH, Bryant D (2006) Application of Phylogenetic Net-
works in Evolutionary Studies. Molecular Biology and Evo-
lution 23: 254–267
Kamdem Toham A, Teugels GG (1997) Patterns of microhabi-
tat use among fourteen abundant fish species from the lower
Ntem River Basin in Cameroon. Aquatic Living Resources 10:
Kamdem Toham A, Teugels GG (1998) Diversity patterns of fish
assemblages in the Lower Ntem River Basin (Cameroon), with
notes on potential effects of deforestation. Archiv für Hydro-
biologie 141: 421–446
Kamdem Toham A, Teugels GG (1999) First data on an index
of Biotic Integrity (IBI) based on fish assemblages for the as-
sessment of the impact of deforestation in a tropical West-
African river. Hydrobiologia 397: 29–38
Kottelat M (1997) European freshwater fishes. An heuristic
checklist of the freshwater fishes of Europe (exclusive of for-
mer USSR), with an introduction for non-systematists and
comments on nomenclature and conservation. Biologia,
Bratislava, Section Zoology 52 (Supplement 5): 1–271
Kullmann H, Klemme B (2007) Female mating preference for
own males on species and population level in Chroma-
phyosemion killifishes (Cyprinodontiformes, Nothobranchi-
idae). Zoology 110: 377–386
Leal ME (2004) The African rain forest during the Last Glacial
Maximum, an archipelago of forests in a sea of grass. PhD
Thesis Wageningen University, The Netherlands
Maley J (1996) The African rain forest – main characteristics of
changes in vegetation and climate from the Upper Cretaceous
to the Quarternary. Proceedings of the Royal Society of Edin-
burgh 104B: 31–73
Meinken H (1951) Aphyosemion cognatum, eine neue Zahnkarp-
fenart aus dem Kongo Gebiet. Die Aquarien und Terrarien
Zeitschrift (DATZ) 4: 255–257
Moritz C, Patton JL, Schneider CJ, Smith TB (2000) Diversifi-
cation of rainforest faunas: an integrated molecular approach.
Annual Reviews in Ecology and Systematics 31: 533–563
Murphy WJ, Collier GE (1999) Phylogenetic Relationships of
African Killifishes in the Genera Aphyosemion and Fundu-
lopanchax inferred from Mitochondrial DNA Sequences. Mo-
lecular Phylogenetics and Evolution 11: 351–360
Myers GS (1924) A new poeciliid fish from the Congo, with re-
marks on funduline genera. American Museum Novitates 116:
Pellegrin J (1904) Cyprinodontidés nouveaux du Congo et de
l’Oubanghi. Bulletin du Muséum National d’Histoire Naturelle
10: 221–223
Plana V (2004) Mechanisms and tempo of evolution in the
African Guineo-Congolian rainforest. Philosophical Transac-
tions of the Royal Society, London, B 359: 1585–1594
Poll M (1951) Notes sur les Cyprinodontidae du Musée du Con-
go belge. Première partie: les Rivulini. Revue de Zoologie et
de Botanique Africaines 45: 157–171
Radda AC, Pürzl E (1987) Colour Atlas of Cyprinodonts of the
Rain Forests of Tropical Africa; Hofmann Verlag, Wien
Scheel JJ (1968) Rivulins of the Old World. TFH Publications,
New Jersey
Scheel JJ (1990) Atlas of the Killifishes of the Old World. TFH
Publications, New Jersey
Seegers L (1997) Killifishes of the World. Old World Killies I.
Aqualog, A.C.S. GmbH, Mörfelden-Walldorf
Sonnenberg R (2007) Description of three new species of the
genus Chromaphyosemion Radda, 1971 (Cyprinodontif-
ormes: Nothobranchiidae) from the coastal plains of
Cameroon with a preliminary review of the Chroma-
phyosemion splendopleure complex. Zootaxa 1591: 1–38
Sonnenberg R, Blum T (2005) Aphyosemion (Mesoaphyosemion)
etsamense (Cyprinodontiformes: Aplocheiloidei: Notho-
branchiidae), a new species from the Monts de Cristal, North-
western Gabon. Bonner zoologische Beiträge 53: 211–220
Sonnenberg R, Blum T, Misof B (2006) Description of a new
Episemion species (Cyprinodontiformes: Nothobranchiidae)
from northern Gabon and southeastern Equatorial Guinea.
Zootaxa 1361: 1–20
Sonnenberg R, Busch E (2010) Description of Callopanchax
sidibei (Nothobranchiidae: Epiplateinae), a new species of kil-
lifish from southwestern Guinea, West Africa. Bonn zoolog-
ical Bulletin 57: 3–14
Swofford DL (1998) PAUP*.Phylogenetic Analysis Using Par-
simony (*and Other Methods).Version 4. Sinauer Associates,
Sunderland, Massachusetts
Tamura K, Dudley J, Nei M, Kumar S (2007) MEGA4: Molec-
ular Evolutionary Genetics Analysis (MEGA) software ver-
sion 4.0. Molecular Biology and Evolution 24: 1596–1599
Thompson JD, Gibson TJ, Plewniak F, Jeanmougin F, Higgins
DG (1997) ClustalX windows interface: flexible strategies for
multiple sequence alignment aided by quality analysis tools.
Nucleic Acids Research 24: 4876–4882
Van Bergeijk WA, Alexander S (1962) Lateral line canal organs
on the head of Fundulus heteroclitus. Journal of Morpholo-
gy 110: 333–346
Van der Zee JR, Huber J-H (2006) Zur Identität von
Aphyosemion christyi (Boulenger, 1915), Aphyosemion
schoutedeni (Boulenger, 1920) und Aphyosemion castaneum
Myers, 1924 (Pisces, Cyprinodontiformes, Aplocheilidae).
Deutsche Killifisch Gemeinschaft Journal 38: 131–142
Van der Zee JR, Sonnenberg R (2010) Aphyosemion teugelsi
(Cyprinodontiformes: Nothobranchiidae), a new species from
a remote locality in the southern Democratic Republic of the
Congo. Zootaxa 2724: 58–68
Van der Zee JR, Woeltjes T, Wildekamp RH (2007) Aplochei-
lidae. Pp. 80–240 in: Stiassny MLJ, Teugels GG, Hopkins CD
(eds.) The Fresh and Brackish Water Fishes of Lower Guinea,
West-Central Africa. Volume II. IRD Éditions, Paris
86 Jouke R. Van der Zee & Rainer Sonnenberg
Bonn zoological Bulletin 60 (1): 73–87 ©ZFMK
Wildekamp RH (1993) A World of Killies. Atlas of the Oviparous
Cyprinodontiform Fishes of the World. Volume 1. The Amer-
ican Killifish Association, Mishawaka, Indiana
Woeltjes T (1984) Aphyosemion lefiniense spec. nov. (Pisces,
Atheriniformes, Rivulineae) aus dem Lefini Subsystem des
Kongo-Flusses in der Volksrepublik Kongo. Deutsche Killi-
fisch Gemeinschaft Journal 16: 3–14
Zhang D-X, Hewitt GM (1996) Nuclear integrations: challenges
for mitochondrial DNA markers. Trends in Ecology and Evo-
lution 11: 247–251
Received: 18.01.2011
Accepted: 04.03.2011
Corresponding editor: F. Herder
A new Aphyosemion from the eastern Congo
Bonn zoological Bulletin 60 (1): 73–87 ©ZFMK
... 13-15). Aphyo semion mengilai and A. grelli are clearly separated from the other known species presenting black margins on unpaired fins in males, Aphyosemion congicum (Ahl 1924) because those species do not present less than 10 dorsal fin-rays, the females do not show a strong reticulation due to dark scale borders, and the edges of the caudal fin in males are not pointed or extended as in the species of the subgenus Aphyosemion (Huber 2005; van der Zee & Sonnenberg 2011). Both species can be distinguished from A. labarrei (Poll 1951) by having stripes on flank (vs. a dense pattern of large red blotches, forming fused irregular bands on caudal peduncle), by the shape of the caudal fin (subtruncated vs. pointed dorsally and ventrally), and by presenting more circumpeduncular scales (15-16 vs. ...
Full-text available
A high concentration of endemic species of seasonal killifishes has been recorded for a small area encompassing the highland plateaus associated with the upper section of the Carnaúba de Dentro River drainage and adjacent drainages of the middle section of the São Francisco River basin, northeastern Brazil. The present study is primarily directed to the taxonomy of the H.fulminantis species complex in this region, and describes habitat decline and extirpation of natural killifish populations recorded in field studies between 1993 and 2017. Both morphological characters and molecular species delimitation methods using single-locus models (GMYC and bPTP) support recognition of two closely related endemic species, H.fulminantis and H.splendissimus Costa, sp. n. The new species is distinguished from other congeners of the H.fulminantis complex by having a red pectoral fin in males, well-developed filamentous rays on the tips of the dorsal and anal fins in adult males, and the second proximal radial of the dorsal fin between the neural spines of the 8 th and 9 th vertebrae in males. Most recent field inventories indicated possible local extinction of populations of H.fulminantis and H.splendissimus in the studied area, but additional field studies should be made in other parts of the upper Carnaíba de Dentro River basin to evaluate the current conservation status of these species.
Full-text available
A great diversity of animal species adapted to life in the semi-arid Caatinga of northeastern Brazil, including seasonal killifishes, has been reported in the last three decades. More recently, field and molecular data have shown a high occurrence of cryptic species. The killifish group herein analysed, the Hypsolebiasmagnificus species complex, is endemic to the middle and southern portion of the Caatinga, occupying about 120 km along the floodplains of the middle São Francisco River and some adjacent tributaries. Species of this complex are rare and presently considered threatened with extinction, being uniquely found in pools protected by trees and bushes. Single-locus delimitation methods were used to test species limits of populations displaying different colour patterns along the whole distribution of the complex. All analyses consistently supported the three nominal species and two new, herein described: H.gardneri Costa, sp. n. , from the floodplains of the middle São Francisco River and H.hamadryades Costa, sp. n. , from the Gorotuba River floodplains. The phylogenetic analysis highly supports H.hamadryades as sister to a clade comprising H.gardneri and H.harmonicus . Our field observations suggest that H.hamadryades is a miniature species. This study indicates that the H.magnificus complex comprises cryptic species apparently endemic to small areas and extremely vulnerable to environmental changes, deserving high concern.
Full-text available
Three new 'Aphyosemion' species are described from the upper Louessé River in the Massif du Chaillu, Republic of the Congo, based on a combination of DNA, habitat preference, male colour pattern, and morphological data. 'Aphyosemion' cyanoflavum, new species, is a member of the 'A'. ogoense group. It differs from its congeners by a unique colour pattern and cephalic sensory system which contains a wide supra-orbital groove with large, densely pigmented anterior neuro-masts and dark frontal neuromasts housed in one pit with one central anterior lobe. 'Aphyosemion' mandoroense, new species , and 'A'. cryptum, new species, are members of the 'A'. coeleste group. 'Aphyosemion' cryptum, new species, is in appearance very similar to 'A'. coeleste, but lacks the typical post opercular metallic green blotch and is generally larger in body size. Initial DNA analyses demonstrate that 'A'. cryptum, new species, is more closely related to 'A'. mandoroense, new species, than to 'A'. coeleste, despite similarity in appearance. 'Aphyosemion' cryptum, new species, and 'A'. coeleste occur syntopic in several locations in a sub-catchment of the upper Louessé system, however differ in microhabitat preference. 'Aphyosemion' mandoroense, new species, differs by male body and fin colour pattern from all species of the 'A'. coeleste group except 'A'. citrineipinnis. From the latter, it can be distinguished by the absence of red pigmentation and a dark grey to black margin in the unpaired fins.
Full-text available
Background. Although the Congo basin is the second largest river basin in the world and it has been considered a biodiversity hotspot for fish, still many parts of this basin remain poorly studied. In this study, we examined the poorly known ichthyofauna of three major north-eastern tributaries of the Congo basin (Itimbiri, Aruwimi and Lindi/Tshopo). A checklist of the ichthyofauna is provided and two synonymies are presented. As such, it contributes to unraveling the poorly known fish diversity in the whole Congo basin. An improved knowledge of the ichthyofauna is a sound baseline for further studies and conservation Material and methods. Fish specimens from five recent expeditions have been identified, and the older collection material from the study area, housed at the Royal Museum for Central Africa and other institutions has been checked and re-identified when necessary. Results. In total, 320 species were recorded, 232 of which from the Itimbiri, 246 from the Aruwimi, and 187 from the Lindi/Tshopo, with the Mormyridae being the most dominant family in all three basins. Micralestes sardina Poll, 1938 is relegated to synonymy with Micralestes humilis Boulenger, 1899, and Enteromius trinotatus (Fowler, 1936) is designated as a replacement name for Enteromius tetraspilus (Pfeffer, 1896). Within the Aruwimi, a clear difference in species richness and composition is apparent between the headwaters (Ituri/Epulu) and the lower reaches near the Congo main stream. The headwaters are characterised by low species richness, with the Cyprinidae being the most dominant family, while the lower reaches are more species rich, with mormyrid species being the most dominant. The presence of two waterfalls on the Ituri/Epulu has a noticeable impact on the fish distribution. Finally, the hypothesis that an ancient connection between the north-eastern part of the Congo basin and the region of the Albertine Rift is still reflected in their present ichthyofauna is not confirmed. Conclusions. This study provides a first checklist of the ichthyofauna in the north-eastern tributaries of the Congo basin, and illustrates the influence of physical barriers on fish diversity and distribution.
Full-text available
Aphyosemion pseudoelegans, new species, is described, based on collections that have been misidentified for decades as A. elegans by all authors. The new species superficially resembles A. elegans, but can be distinguished from the latter and all other species of the genus by a diagnostic combination of colour pattern characters, most prominently it has a dark red to black dorsal fin in males, a colour pattern always absent in all known A. elegans populations, including the type specimens, and an asymmetrical colour pattern on caudal fin margins versus symmetrical in A. elegans. Aphyosemion pseudoelegans, new species, is found in small rivers on the left bank of the middle Congo, where it occurs sympatric and sometimes even syntopic with A. elegans, A. sp. aff. castaneum, or an undescribed species of Aphyosemion at Ikela.
Epiplatys atratus, a new species of the E. multifasciatus group, is described from specimens collected from several tributaries of the middle Lulua River, a tributary of the Kasai River, south of Kananga (Democratic Republic of the Congo, Kasai Occidental Province). Epiplatys atratus is the south-eastern most representative of the genus. Large adult E. atratus males differ from all congeners in displaying a dark grey to black pigmentation of body and fins. In contrast to other Epiplatys species, with a fully exposed laterosensory system of the head, the lobes surrounding the supra-orbital part of the laterosensory system almost completely cover the system in large males of E. atratus. Also in males of E. atratus the dorsal fin is positioned on average more anteriorly than in other members of the E. multifasciatus group. Small males and females show a unique pattern of three fine oblique dark bars just behind the pectoral fin.
Full-text available
Aphyosemion teugelsi is described from specimens collected in a small creek in the upper Wamba River basin in the southwest of the Democratic Republic of the Congo not far from the Angolan border. This is∼400 km outside the known distribution area of the genus. It is distinguished from all other members of the genus Aphyosemion by the combination of broad black margins on all fins in males, a robust lower jaw, a more convex back, a large head with large eyes, a more anterior position of the dorsal fin relative to anal fin, and absence of extensions on the upper and lower caudal fin.
As part of an ongoing project on freshwater fish diversity and conservation of the Lower Ntem River Basin, a synecological study of fish assemblages was undertaken. Two different techniques were used to depict the structural and functional patterns of these fish assemblages: Canonical Correspondence Analysis, an appropriate ordination technique designed for finding the correspondence between physical factors in the environment and biotic factors in the communities, and cluster analysis. Results obtained show a change in both species richness and species composition along the longitudinal gradient. Upstream sites were characterized by a pattern of species addition, while intermediate and downstream sites showed a pattern of species loss and replacement. Measurements of habitat diversity revealed that opposite to depth variability, substrate diversity did not significantly increase downstream. The functional pattern of these fish assemblages almost conformed to the prediction of the River Continuum Concept, with an exception that species richness of invertivores did not increase significantly downstream. A model comprising catchment area, mean % canopy closure, substrate types and chemical variables accounted for 51 % of the variation in community composition. Partial variance explained by forward selected environmental variables highlighted amongst other the importance of the bankside cover (mean % canopy closure) in the model. This statistical insight provide further evidence of the role played b this variable in the structure and function of stream fish communities. With regard to the ongoing deforestation in this area, the implications of these findings for the conservation of freshwater biodiversity are discussed.
Episemion krystallinoron sp. nov., the second species in this formerly monotypic genus, is described from the Monts de Cristal in northwestern Gabon and adjacent areas in Equatorial Guinea. It is clearly distinguished from its congener by the coloration pattern of adult males. Species status is also supported by mitochondrial DNA data. Episemion Radda & Pürzl, 1987 is regarded as a valid genus, distinct from all other nothobranchiid genera and probably most closely related to Diapteron Huber & Seegers, 1977 and Kathetys Huber, 1977. It is the third described endemic nothobranchiid species from the Monts de Cristal and adjacent areas.
A study of microhabitat use by fourteen abundant fish species was undertaken in the lower Ntem River Basin in Cameroon (Africa). Seventy-one microhabitats including pools, raceways and riffles were studied twice during the 1994 and 1995 dry season. Results obtained by Canonical Correspondence Analysis (CCA) showed a good niche separation along the ordination axes I (lambda(1) = 0.37) and II (lambda(2) = 0.20). Microhabitat selection by these species reflects response to several physical factors including width, depth, current velocity and substrate type. However, a substantial part of the variation in species segregation was explained by both instream (% leaves, woods and roots, and aquatic plants) and bankside (% mean canopy closure) cover. A striking pattern encountered during this study was the frequent size-related variation in habitat use displayed by many species. For example, smaller individuals of Brycinus longipinnis, Barbus camptacanthus, Distichodus notospilus, and D. hypostomatus, occurred more often in faster flowing waters than did their larger conspecifics. The association displayed by CCA was high (0.957 and 0.938, respectively for the first and second axis), and Monte Carlo test (99 permutations) revealed that these associations were significantly different from random (p = 0.01). The implications of these results for habitat management are discussed.