courtesy of
ARK - Arizona Rivulin Keepers
The Scheel Letters, No. 48
West African Rivulins: Epiplatys grahami (Boulenger); Draft (Aquarium Journal)
During November 1905, a streamer from West Africa arrived at Hamburg Germany.
This ship brought the first recorded shipment of aquarium fishes to Europe from
this part of the tropical world. The live fishes were taken by W. Schroot and
were soon inspected by amateur zoologist J.P. Arnold who provisionally identified
the four Rivulins contained in the shipment. These Rivulins were caught near Warri
situated at the western part of the Niger River Delta. There were two sorts of
Epiplatys (Haplochilus) that Arnold identified as H. sexfasciatus and H. spilargyreus.
However, he was not able to identify the two species of Aphyosemion (Fundulus),
and they were later on described as F. arnoldi and F. gularis var. blau (=A. coeruleum (now = SJO rjs).
As far as the identification of one species of Haplochilus is concerned, the
E. sexfasciatus was correctly identified. This is easily seen from the descriptions
and pictures of this fish in the aquarium magazines of that time. Arnold's
H. spilargyreus however needs further investigation.
Poecilia spilargyrea was the first West African Rivulin to be described. It was
named by Dumeril in 1861. For this description he had two individuals said to
originate from the Mandingue Coast which corresponds to the coast of Senegambia.
Dorsal 10/anal 14. Dumeril's species may be identical with E. senegalensis or
E. fasciolatus or some Aphyosemion which all live inside this area and which,
like Poecilia spilargyrea, have a rounded and rather short snout. As the type
material in Paris is known to be in a very poor condition and because the description
does not give away sufficient information, Dumeril's species could not at present
be identified with any known West African Rivulin.
The new Epiplatys were soon distributed to German aquarists. A report from a meeting
of the Dresden Aquarium Society of 03 Feb. 1906 says that the well known breeder
of aquarium fish of that time, Mr. Thumm, presented both species to the members
and also reported that he already had offspring form both. Mr. Thumm's fine ink
drawings of aquarium fish of that time are most valuable when one is trying to
identify such old aquarium fish.
During 1909 discussions about the name for the strain of so-called H. spilargyreus
were started in Germany by Thumm. He rejected Arnold's identification and claimed
that the name Haplochilus infrafasciatus should be used. H. infrafasciatus was
described by the English zoologist Guenther back in 1866 and the description was
based on a single individual only. This specimen originated from Old Calabar situated
in the eastern part of the Niger River Delta. In 1915, Boulenger however placed
Guenther's species as a synonym of Epiplatys sexfasciatus Gill. H.S. Clausen has
examined the type specimens of both E. sexfasciatus and H. infrafasciatus and found
that Boulenger was quite right in his idea about Guenther's species. Thumm's idea
did not change the name of the aquarium strain. About 1911 the name H. spilargyreus
was no longer used in German aquarium magazines as the strain probably became
extinct - as it was not a popular aquarium fish.
During 1911 Boulenger, in London, described a new Haplochilus (read Epiplatys)
under the name of H. grahami and he based his description on many specimens collected
by Dr. W.M. Graham in swamps near Lagos in SW Nigeria. In 1912 this name was used
for an aquarium Rivulin in Germany. It is however rather questionable that Boulenger's
new species and the German aquarium fish are identical. After the description of
E. grahami, very few zoological reports of this species were published. In 1933,
Pellegrin described Panchax grahami var. decemfasciata on 3 individuals from Haute
Volta and the northern part of Ivory Coast. If this description is compared with
that of Epiplatys senegalensis Steindachner, Pellegrin's subspecies seems to be a
typical E. senegalensis that is known to live inside the area where Pellegrin's
material originated from. In 1925, Pellegrin reported H. grahami from Ouesso on
the Sangha River. At present it is impossible to verify this identification. We
have been unable to find further reports of E. grahami inside the zoological literature.
Among aquarists the name Epiplatys grahami has been used frequently. A closer
study of the information given about the Rivulins on which this name was placed
makes it quite clear that the aquarists confuse Boulenger's species with Nigerian
strains of Epiplatys sexfasciatus.
During summer 1961 our young friend, Ulf Hannerz from Sweden, made a private expedition
to Nigeria in order to collect fishes studied for the tropical aquarium. Near Benin
City, situated in the western part of the Niger River Delta, he caught several
specimens of Epiplatys that he sent to us for further study. In accordance with
aquarists' tradition he used the name Epiplatys grahami for the whole lot. After
a few days of inspection we were able to divide these Epiplatys into three forms.
Two of these could be identified as E. sexfasciatus and E. longiventralis, whereas
the third form, represented by a single small female, remained unidentified. This
small individual was very much like a young individual of E. sexfasciatus. Only a
close inspection disclosed minor differences. The body was somewhat more slender
and had 6 conspicuous rather broad dark crossbars distributed like the bars of young
E. sexfasciatus. The nature of the crossbars however was not quite like that of
young E. sexfasciatus. Further differences were found in the color of the eye.
Within E. sexfasciatus the eye does not possess much brilliance. The unidentified
specimens however had a strong light green brilliance at the lower part of the eye.
The suspicious specimen was isolated and had an especially heavy feeding and was
soon raised to maturity. During this period a surprising change of the specimen
took place. After maturing the resemblance with E. sexfasciatus was lost. Instead
the female looked very much like a female of E. macrostigma, however it also had
resemblance with E. "chaperi" (see draft on article about E. dageti). The dark
crossbars of (female) E. sexfasciatus tend to disappear on the upper part of the
body sides when the fish is about to mature. Adult female of this species normally
develops rather short dark crossbars which only at the most posterior part of the
body are seen above the midline. The crossbars of the unidentified female grew more
and more narrow and decomposed into vertical rows of dark spots which however were
not reduced in their height and still reached the back. Also the shape of the body
differed more and more from that of female E. sexfasciatus.
During summer 1962, H.S. Clausen caught samples of live Rivulins at many localities
inside Ghana and Nigeria. From receiving young E. sexfasciatus taken in swamps
near Lagos two individuals with shining green eyes were separated. These individuals
were smaller than Ulf's individual and it was impossible to notice any difference
from E. sexfasciatus except for that shining eye. During the growth of these individuals,
the same transformation described for Ulf's specimen occurred. Ulf's female died
soon after the maturing, whereas these two individuals which both were females lived
a long time after maturing, so their whole development could be studied. When
fully matured these females could not be distinguished from females of a probable
Congo Epiplatys (which we had from W. Foersch) which most probably is E. macrostigma
or at least a very closely related form. In order to separate these two strains
one has to handle the fish in such a way that the Lagos females develop their dark
crossbars (-lines) which are not visible on adult females normally. We have had
the probable E. macrostigma female "as egg" from Foersch and dark crossbars have
never been seen on its sides. Both forms have a very thin black line across the
upper third of the anal fin. This line corresponds to the dark longitudinal band
seen in the anal fin of female E. dageti and E. sheljuzhkoi (not on males). Both
have the characteristic marbled pattern of black dots just above the root of the
anal fin.
When the two Lagos females had matured we tried to identify this form with some known
Epiplatys from West Africa. We found that they corresponded very well to Boulenger's
H. grahami, except for a somewhat higher number of dorsal rays (9-10 versus 7). As
we use close up photos for such counting of dorsal rays we normally are able to count
more rays than mentioned in descriptions. The collections of H.S. Clausen of preserved
Epiplatys from southern Nigeria contained numerous individuals belonging to this species
from different localities within the Niger Delta area, from forest swamps inside the
coastal part of western Nigeria. H.S. Clausen also has inspected the type material at
the British Museum and found our material to be identical with H. grahami Boulenger.
The two Lagos females of E. grahami never entered into any natural spawning with any
male of another Epiplatys or with an Aphyosemion. Males and females were not at all
interested, not even when kept together for months. Ulf's young female from Benin
City was much more willing. First it was crossed to a male E. sexfasciatus from
Benin City. Eggs were fertile and the development of the embryos apparently was
normal. However, about the time when the fry should hatch out most embryos died
inside their eggs. Some fry however hatched, but also these fry died after a few
days. It is likely that such hybrids can not be raised.
The next cross was prepared with a male E. "chaperi" (the aquarium strain). The number
of fertile eggs and the development of the embryos was quite normal. The young hybrids
were very viable and some were raised to maturity and reached E. grahami size. They
probably were intersexes with dominant female characters, but also some male characters
(coloration). Dr. Sick prepared electrophoresis spectrals of the blood haemoglobines
and found that the hybrids mentioned above and the Lagos female both had a normal 4
line spectrum just as E. senegalensis, E. fasciolatus, E. dageti and E. "chaperi"
(aquarium). E. sexfasciatus however has a 6 line spectrum.
The E. "macrostigma" mentioned above was imported into Germany in 1963 together with
the so-called Aphyosemion striatum (also known as A. cameronense or A. lujae in
Europe and the USA since 1960), said to originate from the Congo River above Leopoldville.
Foersch sent us some fertile eggs, but as these hatched within the mail we were only
able to save a single specimen. This individual was a female and it never developed
any "sexfasciatus-like" appearance at all. No transverse nor a longitudinal black
band has been seen. H.S. Clausen recently inspected the type material of E. macrostigma
at the British Museum and found that the type specimen had many very thin dark
transverse lines across the body sides and that these lines were just like those on
preserved specimens of E. grahami. However, the form known as an aquarium fish
under the name E. macrostigma does not develop dark crossbars at least on juveniles,
it is likely that this form does not develop such bars. After all we do not think
that the resemblance between E. grahami and E. macrostigma (the zoological species
as well as the aquarium forms) is a matter of convergence. Most probably it indicates
a close relationship. When crossings have not been prepared, it is impossible to
say if these forms represent good species or should be joined into one single species.
J. Lambert from Belgium collected live Rivulins etc. near Lambarene in Gabon (Ogooue
River area) in 1962. He brought home some Epiplatys resembling both E. grahami and
E. macrostigma very much. As no crossbars developed on these specimens they were
referred to E. macrostigma. Bruce Turner in the USA has also sent us pictures showing
some "odd balls" sorted out from shipments said to originate in Nigeria. Also these
individuals are very much like E. macrostigma and E. grahami.
In 1928 in Germany Ahl described a certain Panchax ornatus. His description was based
on aquarium kept specimens without known origin. From the description and the pictures
in aquarium magazines etc. of that time it is quite clear that also this form belongs
to the group mentioned here. As there is no reference to any dark crossbars, not even
on juveniles, this form probably belongs to the aquarium form called E. macrostigma,
or a closely related form. Also a number of the species described by Ahl from West
Africa apparently also belongs to this group, but the very short and insufficiently
detailed descriptions do not make this idea sure. We hope to be able to make this
question clearer when the type material has been inspected.
The information harvested so far for E. grahami and E. macrostigma and their probable
relatives suggest that the former is restricted to a rather narrow area inside the
coastal forest of western Nigeria and the Niger Delta area, whereas the latter seems
to be centered somewhere near the mouth of the Congo River and the coastal areas
up to former French Congo. Inbetween, in the Ogooue River area there is a closely
related form (Lambert's specimens). Ecological information on E. macrostigma is
unknown to us. We know that the adult E. grahami prefers biotopes like forest
swamps and that this species, in this respect, corresponds with E. dageti west of
the Dahomey gap when compared with the chosen biotopes of E. sexfasciatus and E.
sheljuzhkoi respectively. However, we are at present unable to explain why juvenile
individuals of E. grahami prefer to shoal together with juvenile E. sexfasciatus.
Mimicry or protective resemblance???
Finally we have to take one more look on the aquarium Rivulin called H. spilargyreus
by Arnold back in 1905. We are quite sure that this form is identical with Boulenger's
H. grahami, but it is difficult to prove this. Facts that support our idea is that
the strain came from Warri and that juveniles were so much like juveniles of
E. sexfasciatus that it was quite impossible to separate these two forms at that
stage. Several reports are given about the odd transformations of appearance
when the fish matured. Thumm's drawing of that time does not show exactly Boulenger's
form, but the fish pictured is closer to this species than to any other known
Epiplatys. Also the weakly developed sexual difference between male and female
is mentioned which also characterizes Lambert's strain from the Ogooue. The most
conspicuous character of the male is said to be the pointed anal fin.
Live females of E. grahami are reddish brown on the back whereas the belly is nearly
white. The big perfectly rounded spots on the body sides are red, however those
close to the anal fin root are reddish black or even black. Numerous very small
brilliantly red spots are hardly seen by the naked eye on the posterior part of
the dorsal and on the upper part of the caudal fin. There are no red edged scales
on the back as on E. sexfasciatus and many species of Epiplatys. The unpaired
fins are pale yellowish, however the anal fin has a rather powerful yellow color
below the fine black longitudinal stripe across this fin. The hybrid produced
with male E. "chaperi" (aquarium form) often had a broad completely black longitudinal
band from the gill covers to the root of the caudal fin. Only once (photographically
recorded) such a (pale) band has been seen on one of the Lagos females.
The pigmentation of the female's throat is very weakly developed and there is no
conspicuous throat pattern. According to the aquarium literature, the male has a
red lower lip and perhaps a red transverse throat line behind the lower lip. This
means that E. grahami possesses the two basic systems of black body side signals
(longitudinal and transversal bands), whereas the black throat signal system seems
to be absent and is replaced by a permanent system of red lines not unlike that of
Aphyosemion. This is not the only trait pointing towards Aphyosemion, also the whole
appearance of this species (when matured) is rather "Aphyosemion-like".
Our specimens of E. grahami were very stagnant fish and in tanks containing more vivid
fish they tend to hide themselves, mostly near the surface. The two Lagos females
were very hardy fish and gave no problems. They seem to reach a total length of
approximately 45 mm. The egg (Benin strain) is small, 1.0 mm. The egg size thus
corresponds to that of E. dageti, E. macrostigma (Foersch and B. Turner's strains).
The egg membrane has no conspicuous pattern.
