A NEW SPECIES OF 
HEMICHROMIS (PISCES, CICHLIDAE) 
OF SIERRA LEONE AND LIBERIA 
By A. I. PAYNE AND E. TREWAVAS 
INTRODUCTION 
WHILE the junior author (A. I. P.) was making a general survey of the freshwater 
fishes of Sierra Leone, he found a hitherto undescribed species of Hemichromis. 
Some specimens were taken in baited plexiglass traps similar to that described by 
Breder (1960) as a fry-trap ; others were caught by hook and line. 
Among earlier collections from Sierra Leone in the British Museum (Natural 
History) we found that those of Mr N. W. Thomas, received in 1915, and Mr T. S. 
Jones, agricultural officer in the territory in the 19503, also contained examples of 
the new species. The juveniles from Mr Jones had been found sufficiently striking 
to have been labelled with a query as to species. Two from Liberia presented by 
Mr E. Roloff in 1972 and provisionally labelled 'Hemichromis sp/ also belong to the 
new species. 
The description below is based mainly on preserved material, but the colour is 
described from life. 
DESCRIPTION OF NEW SPECIES 
Hemichromis fugax sp. n. 
HOLOTYPE. <J, 72 + 19 mm, from a forest stream, Kassewe Forest Reserve, 
Sierra Leone. 
PARATYPES. Five specimens, 41-76 mm in SL from Kassewe, the Gbangbar 
system near Moyamba and from a small stream with muddy bottom near Njala, as 
listed on p. 166. 
DESCRIPTION. Details of these specimens (except the smallest), two from 
Victoria, Sierra Leone, and one from a stream about 50 km from Monrovia, Liberia 
(coll. Roloff), are given in Table I. The 41 mm paratype, having been preserved 
with the mouth thrust forward, is unsuitable for measurement of proportions. 
Upper profile of head straight, lower jaw slightly projecting. 
Cheek with 3 or 4 horizontal rows of scales, in some i or 2 small scales in addition. 
Teeth of jaws unicuspid, in one row in lower jaw (rarely i or 2 inner teeth), one 
and a short second row in upper jaw, the lengths grading smoothly from longer 
anterior to shorter posterior. 
Gill-rakers short, 2 + 1 + 5-7 on the first arch, the i or 2 lowest often abruptly 
smaller than the others. 
Bull. Br. Mus. nat. Hist. (Zool.) 30, 5 Issued 28 October 1976 
9 
i6o 
A. I. PAYNE AND E. TREWAVAS 
SL (mm) 
As % SL 
Depth 
Head 
TABLE i 
Proportions and some meristic characters in H. fugax 
Holotype 
Kassewe 
72 
Paratypes 
Kassewe 
76-5 66-5 
Moyamba 
70 
35'7 
39-3 
12-2 
9'45 
16-5 
11-7 
16-8 
25-0 
31-0 
22-5 
12-7 
24-0 
33-o 
41-8 
26 
3i 
XIV 1 1 
Njala 
43 
Victoria 
80 53'5 
Liberia 
59 
sn 
int 
l.j. 
c.p.l. 
c.p.d. 
P 
As % head 
sn 
eye 
pr 
int 
u.j. 
l.j. 
Sc.1.1. 
Sc.D-1.1. 
D 
A (III + ) 
Head = length of head ; sn = snout, from middle of upper lip to edge of orbit ; pr = depth of pre- 
orbital bone ; int = interorbital width, the roof of the skull between the eyes ; u.j. = upper jaw, from 
middle of upper lip to end of maxillary ; l.j = lower jaw, from middle of lower lip to end of retro- 
articular ; c.p.l. and c.p.d. = length and depth of caudal peduncle ; P = length of pectoral fin ; Sc.1.1. 
= no. of scales in lateral line series, resuming at the end of the upper lateral line with the lower lateral 
line scale (pierced or not) in the oblique row behind, the row used sloping downwards and forwards ; 
D = dorsal fin-rays ; A = soft anal rays. SL = Standard Length. 
XIV ii XIII 12 XIII 12 XIV ii 
2mm 
FIG. i. The lower pharyngeal bones of H. bimaculatus (left) and H. fugax (right). 
NEW SPECIES OF HEMICHROMIS 161 
Scales cycloid with gothic circuli, 25 to 27 in the lateral line series, 3, 3^ or 4 from 
origin of dorsal to lateral line. 
Caudal fin rounded. Length of caudal peduncle 0-75-0-9 of its depth. 
Lower pharyngeal bone hastate (Fig. i) with rather long lateral apophyses. 
Teeth slender, but the posterior ones of the middle rows may be worn down ; lateral 
posterior and several others with a crest of 1-4 small brown points in front of the 
main cusp. In the Liberian specimen some teeth with crest not divided into points, 
others with one, two or three in addition to the main cusp ; anterior teeth unicuspid. 
Teeth of posterior row 16-20. 
Colour in life (taken straight from the water). General appearance olive brown 
with three approximately equidistant black blotches on each side, the first on 
operculum, the third at base of caudal ; much red on the fins. 
Head with variable gold and red streaks on cheek and gill-cover and a greenish 
gold rim to the opercular spot. A circumpupillar gold ring, rest of iris grey or (in 
two of four observed) with red lower quadrant. Lips and chin sometimes tinged 
red. Lappets of dorsal fin, apex of soft fin and often also its upper edge bright 
red ; a clear submarginal band ; rest of fin membrane golden-yellow to pinkish 
with clear or pale blue spots. Anal fin clear proximally, reddish distally. Caudal 
fin with distal quarter to three-quarters bright red, proximally yellowish ; a clear 
submarginal streak underlining the red upper corner. Pectoral transparent, 
colourless or yellow. Pel vies clear with red or black leading edge. 
After a minute or two out of water small pale turquoise spots may appear on the 
flank, but these are much smaller and less iridescent than those of H. bimaculatus 
Gill. With time reddish centres to the scales may also appear. 
After two days in a tank one fish became much more sombre and the middle 
flank spot disappeared. A dark bar through the eye became pronounced and eight 
faint vertical bars could be switched on. 
Preserved specimens show the three lateral blotches and often traces of the 
vertical bars. 
DIAGNOSTIC FEATURES. Distinguished from H. bimaculatus by the long, acute 
snout and narrow interorbital space, giving the ratio of Table 2 (p. 164) and from 
H. fasciatus by lower numbers of scales (29-30 in H. fasciatus] and the fact that the 
anterior teeth are not abruptly contrasted in length with the rest but grade smoothly 
into the series. 
ECOLOGY. The fish has been caught in fairly fast-flowing forest streams, where it 
tends to lie in the lee of weed-beds, stones or logs, and also in swampy areas. In 
five localities sampled it was accompanied by H. thomasi (Boulenger) (three places) 
or H. fasciatus Peters (two places), but not by H. bimaculatus. Although both 
species were sent by N. W. Thomas in 1915 from 'Victoria' we have no details of the 
locality or localities in which they were taken. 
DISTRIBUTION. The sites where H. fugax has been found in Sierra Leone, along 
with the records for H. bimaculatus, are indicated in Fig. 3. H. fugax appears to be 
restricted to the river systems east of the Rokel and the Freetown Peninsula. 
Samples from the most easterly rivers of Sierra Leone, Moa and Mano, have not yet 
162 A. I. PAYNE AND E. TREWAVAS 
been obtained, but some juveniles described below were taken not far west of River 
Moa and no doubt it will be found in suitable places between here and the Liberian 
locality. 
H. bimaculatus also occurs in the east of Sierra Leone but in the territory so far 
it has been located only in the physiologically more demanding brackish water areas 
of Sherbro Island and the coastal region opposite, including Lake Kwarko, a 
lagoon in the waterways parallel to the coast into which Rivers Waanje and Sewa 
drain. H. bimaculatus may therefore be physiologically more adaptable than 
H . fugax. It is widespread in other African territories. 
The Liberian specimen of H. fugax mentioned above and a smaller one collected 
by Herr Roloff at the same time show that the range of H. fugax must extend 
eastwards, but the limits have yet to be determined. We have not found it among 
the material from the rest of Africa in the BMNH. 
JUVENILES OF THE SPECIES OF HEMICHROMIS 
Young H. fugax of about 15 mm SL obtained at Kassewe have a black median 
longitudinal stripe from the operculum to the end of the caudal peduncle, in 
preserved specimens ending in a slightly darker blotch at the base of the caudal fin. 
Seven young of 21-0-22-5 mm SL from Potoru (a few kilometres west of the 
expanded part of River Moa), preserved in the BMNH, show this broad band 
expanded at the position of the mid-lateral blotch. At the base of the caudal fin 
it ends as an intense black blotch, narrowly extended on the middle caudal rays to 
the tip of the fin (Fig. 2). These have the acute snout and triradiate pharyngeal 
bone characteristic of H. fugax, in contrast to young of H. bimaculatus of the same 
size. 
Young of H. bimaculatus from Bonthe, Sherbro Island (where adults have also 
been collected), show in contrast separate opercular, mid-lateral and caudal blotches 
with no extension on to the caudal rays (Fig. 2). They have the more robust 
pharyngeal bone characteristic of this species. In Lake Kwarko (or Kwako) the 
young are similar. 
The difference in the pigmentation of the caudal between the two samples of 
H. fugax suggests that this pattern may not be strictly specific ; or possibly the fact 
that those lacking the caudal streak were smaller may be significant. In H. 
bimaculatus too there is some variation. In a sample of thirteen, 15-5-44-5 mm in 
SL, from Kiyawa River near Katagum, Nigeria, the two smallest have the caudal 
blotch produced a short distance on the caudal fin and between the mid-lateral and 
caudal blotches is a dusky band. The same is true of a 20-5 mm juvenile from 
about 19 km north of Monrovia taken with two adult H. bimacidatus. It has the 
appearance of H. bimaculatus, but the pharyngeal bone is equivocal. (This sample 
suggests that a borderline between the distribution of H. fugax and H. bimaculatus 
may lie between 20 and 50 km from Monrovia.) 
Some fry of H . fasciatus from Sherbro Island have the lateral band emphasized 
at the positions of the future lateral blotches and continued on the middle caudal 
rays. This caudal extension was not figured for young of this species at Yangambi, 
NEW SPECIES OF HEMICHROMIS 
FIG. 2. Juveniles of (above) H. fugax from Potoru and (below) H. bimaculatus from 
Bonthe with their pharyngeal bones. 
Congo, by Gosse (1963, fig. 17), but the two 'forms', A and B, of H. fasciatus are 
found in Sierra Leone and the fry may give a clue to their distribution there. 
A recent account of the breeding and rearing of young of H. fasciatus from 
Liberia (Nieuwenhuizen, 1975) shows, among other handsome photographs, a shoal 
of 3-week-old fry. Their colour-pattern is almost exactly that of our fry from 
Potoru (Fig. 2), but with a more pronounced upper stripe on each side of the base 
of the dorsal fin. The identity of our juveniles with H. fugax has been confirmed 
by the meristic characters (scales in lateral line series 26-27, against 29-30 in 
H. fasciatus, both 'forms') and that of Nieuwenhuizen's fry by the photograph of 
the parent, which has five black blotches along the side behind the opercular spot. 
164 A. I. PAYNE AND E. TREWAVAS 
This is probably the 'B form' of H.fasciatus (? =H. elongatus Guichenot in Dumeril, 
1859 cf. Loiselle, quoted in Trewavas, 1974) since the 'A form' is notoriously so 
aggressive that it is almost impossible to rear it in an aquarium. 
RELATIONSHIPS OF H. FUGAX 
In meristic characters this species cannot be distinguished from H. bimaculatus 
Gill, also abundant in Sierra Leone, but it is easily recognized by the different colour- 
ing, especially the lack (or lower brilliance) of the turquoise spots on the flanks, and 
by the acute snout (PI. i and Fig. 2). The difference in snout length between the 
two species (Table 2 and Fig. 3) is slight, with some overlap, but the interorbital 
TABLE 2 
Length of snout (Sn) and interorbital width (Int) in Hemichromis fugax and H. bimaculatus 
and the ratio of snout to interorbital width 
H. fugax (N = 13) H. bimaculatus (N = 19) 
Range Mean Range Mean 
SL (mm) 40-7 -80-5 61-21 44-0 -100-0 61-45 
Sn (% head) 26-0 -30-6 28-33 22-6 - 28-6 25-55 
(% SL) 9-5 -11-9 10-88 8-45- 10-2 9-27 
Int (% head) 20-0 -23-0 21-20 22-2 - 26-8 24-73 
(% SL) 7-6 - 8-95 8-18 8-2 - 9-85 8-96 
Sn/Int 1-23- 1-50 1-33 0-86- 1-19 1-04 
width is less in H. fugax and the snout /inter orbital ratio consequently gives a 
reliable difference between the two species. They are also distinguished by the 
pharyngeal teeth, fewer and more slender in H . fugax, sometimes with up to four 
minor cusps, whereas in H. bimaculatus some of the median teeth are stout and 
blunt. In both adult and juvenile the pharyngeal bone is more robust in H. 
bimaculatus. The teeth of the jaws are variable in both species, an inner series being 
present or absent, when present of a few teeth only. In neither species are the 
anterior so sharply contrasted in size with the others as in H. fasciatus Peters, but 
the shape of the snout and mouth in H. fugax approaches that in H. fasciatus. 
The significance of the difference in colour-pattern between our samples of 
juveniles can only be judged after tests of variation between populations and 
developmental stages. Possibly the juvenile pattern is another feature in which 
H. fugax resembles H. fasciatus and differs from H. bimaculatus. 
We have mentioned H. thomasi as though it were a species of Hemichromis, as 
Loiselle & Welcomme have treated it (1972). This is, in any case, an isolated species 
not liable to be confused with any of the currently recognized species of Hemichromis 
and certainly not with H. fugax, having a very short, blunt snout and smaller, more 
numerous pharyngeal teeth. 
We have also examined syntypes (or topotypes) of H. letourneuxi Sauvage 
(i88oa) from Lake Mareotis, Nile delta ; of H. rolandi Sauvage (1881) from Zibane, 
NEW SPECIES OF HEMICHROMIS 
165 
10 
10 
o 
A 
O 
SNOUT 
50 
S L (mm) 
60 
70 
80 
90 
100 
O 
o 
O 
q A 
INTERORB. 
FIG. 3. Comparison between H.fugax (A) and H. bimaculatus (O) in length of snout and 
interorbital width (in mm). 
Constantine, Algeria ; H. saharae Sauvage (i88ob) from near Touggourt ; and the 
large specimens that are the syntypes of H. guttatus Giinther, of unknown locality. 
These names have been regarded as synonyms of H. bimaculatus (by Pellegrin, 
1904 : 219, and Boulenger, 1915 : 431) and we are satisfied that H. fugax is distinct 
from them all. 
ACKNOWLEDGEMENTS 
We are grateful to Mr Paul L. Loiselle for communicating to us his views on the 
species of Hemichromis and particularly his opinion that the type of H. bimaculatus, 
which he has seen, is the species to which we have attributed the name. We also 
thank Dr G. Corbet and Dr P. H. Greenwood for reading the script and suggesting 
improvements. 
1 66 
A. I. PAYNE AND E. TREWAVAS 
MATERIAL EXAMINED 
Hemichromis fugax SL (mm) 
BMNH 1976.1.28.1 (holotype) 72 
BMNH 1976.1.28.6-7 (paratypes) 66, 76 
BMNH 1976.1.28.2-3 (paratypes) 41, 70 
BMNH 1971.8.13.24 (paratype) 43 
BMNH 1976.1.28.4-5 45, 46-5 
BMNH 1976.1.28.8 72 
BMNH 1976.1.28.9 53 
BMNH 1976.1.28.10-13 52-65 
BMNH 1915.5.27.21-23 40-80 
BMNH 1958.9.18.224-9 21-22-5 
BMNH 1972.3.16.11-12 28, 59 
Hemichromis bimaculatus 
BMNH 1888.10.19.23-32 24-74-5 
BMNH 1899.11.25.4 100 
BMNH 1915.5.19.28-30 4 8 -5-72 
BMNH 1915.5.27.24 63 
BMNH 1950.9.22.50-70 22-5-71 
BMNH 1958.9.18.219-222 45'5-89 
BMNH 1958.9.18.223 25-3 
BMNH 1958.9.18.230-239 22-36 
BMNH 1958.9.18.240-1 27, 33 
BMNH 1969.11.19.20-21 (3 fishes) 20-5, 62-5, 6 
BMNH 1928.7.3.112-116 2 9'5~45'5 
BMNH 1930.3.22.276-284 (13 fishes) 15-5-44-5 
BMNH 1884.5.2.3-4 (syntypes of 39'5, 41*0 
H. saharae) 
BMNH 1884.5.2.5-7 (syntypes of 41-2-44-0 
H. rolandi) 
BMNH 1898.2.15.1 (syntypes of 47'5> 5 1>o 
H. letourneuxi) 
BMNH 1860.4.19.3-4 (syntypes of 90-7, 95-5 
H. guttatus) 
Locality 
Forest stream, Kassewe 
Forest Reserve, S.L. 
Forest stream, Kassewe 
Forest Reserve, S.L. 
Gbangbar system nr 
Moyamba, S.L. 
Stream nr Njala, S.L. 
nr Moyamba, S.L. 
River Taia, nr Njala, 
S.L. 
Kassewe, S.L. 
River Tabe, S.L. 
'Victoria, S.L.' 
Potoru, S.L. 
50 km from Monrovia, 
Liberia 
Freetown, S.L. 
Mountain stream, 
Freetown, S.L. 
North Sherbro District, 
S.L. 
'Victoria, S.L.' 
Lake Kwarko, S.L. 
(7i8'N, n59'W) 
Black 'River', Sherbro 
Is., S.L. 
Bonthe, S.L. 
Bonthe, S.L. 
RoKupr, S.L. 
19 km from Monrovia, 
Liberia 
Kiyawa River, nr 
Katagum, Nigeria 
Kiyawa River, nr 
Katagum, Nigeria 
'Oued Rhir Constantine' 
'Oued Rhir, Constantine, 
Algeria' 
Lake Mareotis, Egypt 
Collector 
A. I. Payne 
A. I. Payne 
A. I. Payne 
A. I. Payne 
A. I. Payne 
A. I. Payne 
A. I. Payne 
A. I. Payne 
N. W. Thomas 
T. S. Jones 
E. Roloff 
R. Dinzey 
Hopkins 
N. W. Thomas 
N. W. Thomas 
T. S. Jones 
T. S. Jones 
T. S. Jones 
T. S. Jones 
T. S. Jones 
C. Steiner 
LI. Lloyd 
LI. Lloyd 
Mus. Hist. 
Nat. Paris 
Mus. Hist. 
Nat. Paris 
Mus. Hist. 
Nat. Paris 
Stevens (pur- 
chased) 
NOTE ON THE LOCALITIES 
Most of the localities mentioned can be found on a modern map and by reference 
to our Fig. 4, but with two we had difficulty. 
'Victoria, Sierra Leone.' The Official Standard Names Gazetteer gives the 
position of three such places, but in the maps consulted by us only those of an early 
Handbook of Sierra Leone (Goddard, 1925) mark any of them, and this only one, 
NEW SPECIES OF HEMICHROMIS 
167 
nearest to the Gazetteer reference 739'N, i22o'W. This is a very small place a 
short way inland at the latitude of the north end of Sherbro Island, north of the 
delta of River Jong. Among the other fishes sent from this address was one 
'Psettus sebae' , which confirms it as a locality near the coast. 
Potoru. This also has three references in the Gazetteer, but the probable one is 
some 8 km west of River Moa at its expanded part. This is the only one marked 
FIG. 4. The site records of H. fugax (A) and H. bimaculatus () in Sierra Leone. At 
one site (?) fishes that were probably H. bimaculatus were observed but no specimens 
were taken. 
168 A. I. PAYNE AND E. TREWAVAS 
on a map of 1950 (Director of Colonial Surveys, London), the period when Mr Jones 
was resident in Sierra Leone. It is also marked on the American i : i ooo ooo map 
NB 29 ser. 1301. 
'Oued Rhir, Constantine.' This is the locality given in the register of BMNH 
for specimens of both H. saharae and H. rolandi. The original descriptions give 
for H. saharae 'Sahara, aux environs de Touggourth' and for H. rolandi 'Zibans, 
Sahara, Province de Constantine.' Both these localities are probably in the 
catchment area of Oued Rhir and Chott Melrhir. These and the specimens of 
H. letourneuxi in the BMNH are listed as 'types' by Boulenger (1915). Other 
syntypes remain in Paris. 
REFERENCES 
BOULENGER, G. A. 1915. Catalogue of African freshwater fishes. Vol. III. London. 
BREDER, C. M. 1960. Design of a fry trap. Zoologica N.Y. 45 : 155-159. 
DUMERIL, A. H. A. 1859 Reptiles et poissons de 1'Afrique occidentale. Arch. Mus. nat. 
Hist. Paris 10 : 137-268, pis xiii-xxiii. 
GILL, T. 1862. On the West Africa genus Hemichromis and descriptions of new species in 
the museums of the Academy and Smithsonian Institutions. Proc. Acad. nat. Sci. Philad. 
1862 : 134-139. 
GODDARD, T. N. 1925. Handbook of Sierra Leone. London. 
GOSSE, J. P. 1963. Le milieu aquatique et 1'ecologie des poissons dans la region de Yangambi. 
Ann. Mus. r. Afr. centr. no. 116 : 113-249, pis i-x. 
GUNTHER, A. 1862. Catalogue of the Acanthopterygii Pharyngognathi in the collection of the 
British Museum. Vol. IV. London. 
LOISELLE, P. L. & WELCOMME, R. L. 1972. Description of a new genus of cichlid fish from 
West Africa. Rev. Zool. Bot. afr. 85 : 37-57. 
NIEUWENHUIZEN, A. VAN DEN, 1975. Hemichromis fasciatus - gouden schoonheid uit Afrika. 
II. Het Aquarium Jg. 46 : 30-34, 7 figs. 
Official Standard names Gazetteer No. 101. Sierra Leone. U.S. Board on Geographical Names. 
PELLEGRIN, J. 1904. Contribution a 1'etude anatomique, biologique et taxonomique des 
poissons de la famille des cichlides. Mem. Soc. zool. Fr. 16 : 41-401, pis iv-vii. 
PETERS, W. C. H. 1857. Neue Chromiden Gattung. Mber. K. preuss. Akad. Wiss. 1857 : 403. 
SAUVAGE, E. i88oa. Notes sur quelques poissons recueillis par M. Letourneux en Epire, a 
Corfu et dans le Lac Mareotis. Bull. Soc. philom. Paris (7) 4 : 211-215. 
i88ob. Description de quelques poissons d'especes nouvelles dans la collection du Musee 
d'Histoire Naturelle. t.c. 220-228. 
1 88 1. Description de quelques poissons d'especes nouvelles dans la collection du Musee 
d'Histoire Naturelle. Bull. Soc. philom. Paris (7) 5 : 101-107. 
TREWAVAS, E. 1974. The freshwater fishes of Rivers Mungo and Meme and Lakes Kotto, 
Mboandong and Soden, West Cameroon. Bull. Br. Mus. nat. Nist. (Zool.) 26 : 331-419, 
4 pis. 
DR A. I. PAYNE 
LANCHESTER POLYTECHNIC 
PRIORY STREET 
COVENTRY 
DR E. TREWAVAS 
Department of Zoology 
BRITISH MUSEUM (NATURAL HISTORY) 
CROMWELL ROAD 
LONDON SW7 5BD