100

CICHLIDS yearbook

Ad Konings (Ed.)

Volume 2The

4

1

THECICHLIDS

YEARBOOKVolume 2

Ad Konings (Editor)

2

Cover photographs:1 - Protomelas sp. “Steveni taiwan”, Lake Malawi, Malawi.2 - Tilapia tholloni, Alima River, Congo.3 - “Cichlasoma” sp. “Labridens Tamasopo”, Pánuco, Mexico.4 - Tropheus moorii (Murago), Lusingo, Lake Tanganyika, Zaïre.

Text and photographs by Ad Koningsexcept as otherwise indicated

Mary Bailey (Crediton, UK)corrected the manuscript

The editor wants to thank the following persons whosupplied various cichlids for photographic purposes:Peter Baasch (Stegen, Germany)Marc Danhieux (Maltavi, Hohenahr-Erda, Germany)Alain Gilot (Zaïre Cichlids, Kalemie, Zaïre)

Stuart Grant (Salima, Malawi)

René Krüter (Krüter Tropicals, Rotterdam, Netherlands)Roland Numrich (Mimbon Aquarium, Köln, Germany)Edwin Reitz (Aquapport, Ronnenberg, Germany)Dirk Verduijn (Verduyn Cichlids, Zevenhuizen, Netherlands)

1

2 3 4

Distributors:

USA: Old World Exotic Fish, Inc., P.O.Box 970583, Miami, Florida 33197UK: Animal House (U. K.), Ltd., QBM Business Park, Birstall, Batley, West Yorkshire WF17 9QDSweden: Fohrman Aquaristik AB, Odds Väg 7, 433 75 PartilleGermany: Aquapport (Edwin Reitz), Köselstraße 20, 3003 RonnenbergNetherlands: NVC, Lieshoutseweg 31, 5708 CW Stiphout

ISBN 3-928457-05-5

Copyright © 1992. Cichlid Press. All rights reserved.No part of this publication may be reproduced, stored in a retrieval system, or transmitted in any form or by any means—electronic, mechanical, photocopying, recording or otherwise—without the prior permission of the authors, and the pub-lisher.

Cichlid Press, Blütenweg 17, 6837 St. Leon-Rot, Germany

Printed by RAKET B.V., Pijnacker, Holland

3

The CICHLIDS yearbookCONTENTS

Introduction ...........................................................................................................................................................5

Tanganyikan CichlidsClues to a step-wise speciation by Ad Konings ....................................................................................................6Spawning Xenotilapia spiloptera by Ad Konings ..............................................................................................10The genus Tropheus by Ad Konings ...................................................................................................................13Simochromis marginatus........................................ 16Simochromis babaulti ............................................ 17Xenotilapia ornatipinnis ........................................ 18Ctenochromis benthicola ....................................... 19Neolamprologus petricola ..................................... 20Neolamprologus leloupi ......................................... 21

Neolamprologus sp. “Ubwari Buescheri” ............. 22Neolamprologus sp. “Kavalla” .............................. 23Cyphotilapia frontosa ............................................ 24Cyprichromis sp. “Leptosoma Jumbo”.................. 26Paracyprichromis brieni ........................................ 29

Malawian CichlidsThe genus Cynotilapia Regan, 1922 by Andreas Spreinat ................................................................................30The Protomelas taeniolatus-complex by Ad Konings ........................................................................................34Tramitichromis lituris by Ad Konings ................................................................................................................38Stigmatochromis sp. “Modestus Makokola” ......... 41Dimidiochromis strigatus ...................................... 42Copadichromis boadzulu ....................................... 43Buccochromis rhoadesii ......................................... 44Corematodus taeniatus .......................................... 45Aulonocara rostratum ............................................ 46

Taeniochromis holotaenia ...................................... 48Lethrinops micrentodon ......................................... 50Labeotropheus fuelleborni ..................................... 51Pseudotropheus saulosi ......................................... 52Labidochromis sp. “Hongi” ................................... 53

Victorian CichlidsIntroduction to taxonomy and ecology: Part I by Ole Seehausen ......................................................................54

West African CichlidsLamprologus sp. “Kinganga” by Frank Warzel .................................................................................................60Tilapia tholloni (Sauvage, 1884) by Jan ‘t Hooft ...............................................................................................62Ctenochromis polli / Thoracochromis demeusii by Martin Geerts ...................................................................64

Central American CichlidsThe “Cichlasoma” labridens-complex by Juan Miguel Artigas Azas ...............................................................65“Cichlasoma” minckleyi Kornfield & Taylor, 1983 by Ad Konings ..................................................................71“Cichlasoma” septemfasciatum ............................. 74Theraps coeruleus .................................................. 75

Thorichthys pasionis .............................................. 76

South American CichlidsCrenicichla species from the Rio Xingú by Frank Warzel ................................................................................77Crenicichla sp. cf. regani ....................................... 82Acaronia vultuosa .................................................. 83

Guianacara sp. “Red Cheek” ................................ 84Krobia species ........................................................ 85

Cichlid MaintenanceThe “Aufwuchs-feeder” by Roger Häggström...................................................................................................86Breeding Tropheus the natural way by Gerard Tijsseling ..................................................................................87

Cichlid LiteratimThe cichlid bible by Martin Geerts ....................................................................................................................88Cichlid classics by Mary Bailey .........................................................................................................................90Spawning techniques in mouthbrooders by Ethelwynn Trewavas & Ad Konings .............................................93

4

5

No less than eighteen authors have put together thesecond volume of the cichlids yearbook. It was pleas-ing that almost all contributors to the first volume wereable to write down another piece of their knowledgeand experience in this one. In addition eight new au-thors have written for this volume. Another pleasingfact is that Horst Walter Dieckhoff (Lake Fish, Herten,Germany), the famous underwater photographer, isback in “cichlid-business”. For this volume he has pro-vided some of his photographs but we have made someplans for the near future....

The “new” authors are briefly introduced in thesame order as their articles appear:

Hans-Joachim Herrmann (Hamburg, Germany), aTanganyika specialist and known for his book on thesecichlids, describes two new variants of Simochromis,which he has collected himself.

Another book-author and Malawi specialist, Dr.Andreas Spreinat (Göttingen, Germany), deals withthe genus Cynotilapia and relates his experience withsome other cichlids as well.

Edwin Reitz (Aquapport, Ronnenberg, Germany),a professional and very experienced cichlid breeder,describes the breeding procedure of Aulonocararostratum.

Recently the first commercial shipments of fish fromLake Nyasa, as Lake Malawi is known in Tanzania,have been exported from that country. Two independ-ent operations have been started. In the next volumewe will have some more species for you but this timePeter Knabe (Schlangen, Germany), who is a scien-tist working in Tanzania in the Haplochromis Ecol-ogy Survey Team and thus has first-hand experience,has contributed his first instalment.

Jan ‘t Hooft (Hendrik Ido Ambacht, Netherlands),a lifetime aquarist and the founder of the Dutch cichlidassociation, tells us about one of his passions, namelyTilapia.

Do you keep mbuna with larger Malawian haplo-chromines in your tank? Then you know that at feed-ing time the mbuna devour the major part of the food.It usually takes them ten seconds to do so. RogerHäggström (Örnsköldsvik, Sweden), the editor ofCiklid Bladet, the periodical of the Swedish cichlidassociation, tells us how we can keep the mbuna busyeating.

The CICHLIDS yearbook

Last but not at all the least in the long row of “new”authors, Mary Bailey (Crediton, UK) explains scien-tists (should) give names to fish. Mary, who has beenon the committee of the British Cichlid Associationsince 1982 and has studied English and Latin, wasalso of tremendous help in correcting the manuscriptsof the yearbooks. She has kept and bred many speciesand is an aquaristic consultant for several aquariummagazines in the UK.

A very important publication has been issued re-cently, namely Cichlid Fishes; Behaviour, Ecology andEvolution. It is the cichlid bible for the years to comefor every aquarist interested in cichlids. It is discussedby Martin Geerts.

In the previous volume I have reported onNeolamprologus leloupi and said that N. caudo-punctatus should be regarded as a synonym of thisspecies. However, when I visited the locality in Zaïrewhere the yellow-dorsalled cichlid is collected, it wasimmediately clear that I was wrong: both species, N.caudopunctatus and N. leloupi , live sympatrically inthe same habitat. I further explain the situation on page21.

With regard to Willem Heijns’ contribution in vol-ume 1 about “Cichlasoma” spinosissimum , Jaap-Jande Greef (Parrish, Florida), who rediscovered this spe-cies for the hobby, wrote me that he found the speciesin pools at the southern end of Lago de Izabal but notin the lake itself. Jaap-Jan collects fish in CentralAmerica and Africa and breeds them for a hobby.

Without the hospitality and cooperation of severalexporters and friends it would not have been possibleto show you the cichlids in their natural habitat andgive information on how they live. I therefore grate-fully acknowledge Stuart M. Grant (Salima, Malawi),Alain Gillot (Zaïre Cichlids, Kalemie, Zaïre), GaryKratochvil (San Antonio, Texas), Juan Miguel ArtigasAzas (San Luis Potosí, Mexico) and Mireille Schreyen(Fishes of Burundi, Burundi).

Ad Konings

Introduction

6

Among the many endemic species of cichlids in LakeTanganyika which are known to date, there are anumber of species which have evolved geographicalraces. The best known examples of such cichlids arethe species of the Tropheus-complex. Many differentgeographical variants have been found which all seemto belong to five, maybe six, species. Besides Tropheusthere are other species like Cyathopharynx furcifer andOphthalmotilapia ventralis which have evolved sev-eral different races. Not only the mouthbrooders amongthe cichlids have formed distinct populations but also

TANGANYIKANCICHLIDS

Clues to a step-wise speciation

Ad Konings

This race of Ophthalmotilapia ventralis at M’Toto, Zaïre closely resembles that from Cape Mpimbwe, Tanzania (see page 7).

substrate-spawners like Neolamprologus caudo-punctatus and N. leloupi (for the latter species see page21).

Recently, Martin Geerts and I made a short expedi-tion along the southern Zaïrean coast from Kalemieto Moliro. It was our intention to make a preliminaryinventory of the rock-dwelling cichlids in this part ofthe lake. During our trip we became aware of a re-markable feature which may have a greater impact onthe theory of speciation than our first observationsindicated. We observed geographically isolated

7

Ophthalmotilapia ventralis at Cape Mpimbwe, Tanzania.

populations of species whose coloration had a remark-able resemblance to populations along the oppositeside of the lake. We knew about these populationsthrough the many photographs made by WalterDieckhoff in the Tanzanian part of the lake.

Ophthalmotilapia ventralis inhabits the southernhalf of the lake and is known in about a dozen distin-guishable colour varieties, each geographically iso-lated. On the Tanzanian coast at Cape Mpimbwe livesa race which is characterized by a black-blue groundcolour with an irregular, broad white band running di-agonally across the body (photo this page).

Between Cape Tembwe and the bay of Zongwe on

the Zaïrean coast an almost identical variant of O. ven-tralis is found (see photo page 6). Further south alongthe coast a yellow race of O. ventralis inhabits theshallow rocky biotope. The race found at Kapampaclosely resembles that found at Kipili, Tanzania(Konings, 1988: 61). At Lupota, Zaïre, the variantshows more yellow pigment and is almost identicalwith that found at Malasa Island, Tanzania.

Cyathopharynx furcifer is known from most loca-tions around the lake and about six different ’ racesare known. The race found along most of the southern

part of the Zaïrean coast (see photo page 8 top) is prac-tically identical to the one on the other side of the lake(see photo page 8 bottom). In the Zambian part, how-ever, another race inhabits the intermediate and sandybiotopes (see photo page 89).

The geographical variant of T. moorii found betweenMoba and Zongwe closely resembles the small popu-lation at Kibwesa, Tanzania (see Konings, 1990 TFH39 (3): 71) while the race found at Kapampa (see photopage 9) shows a strong resemblance to the race nearMpulungu. The latter two populations, however, areseparated by a long shoreline which is inhabited byabout a dozen different other variants.

T. polli has been described from the central Tanza-nian coast and is characterized by having four spinesin the anal fin, a feature it shares, among its conge-ners, only with T. annectens. Other Tropheus have 5to 7 spines. At M ’Toto, Zaïre, I observed T. annectensin its habitat and found no apparent difference in be-haviour or morphology to that of T. polli.

Neolamprologus leloupi occurs on both sides of thelake (in the southern half) but is not found in Zambia.On page 21 some of its affinities with N. caudo-punctatus are discussed. Both at Kapampa, Zaïre (see

8

Cyathopharynx furcifer,a male photographed south of Moba, Zaïre.

Cyathopharynx furcifer at Kipili, Tanzania.

9

Tropheus moorii at Kapampa, Zaïre.

photo page 21) and at Kalambo, Tanzania/Zambia(René Krüter, pers. comm.) an identical race of N.caudopunctatus – which is not known to occur any-where else — lives sympatrically with N. leloupi atthe borderline of both species’ distribution.

There are several other species which have similar(not identical as in the previous examples) popula-tions on both sides of the lake, e.g. N. sexfasciatus,several species of the N. brichardi-complex andChalinochromis sp. “Popelini” in Zaïre and C. sp.“Bifrenatus” in Tanzania. With regard to the follow-ing discussion, it is important to realize that differentraces of the species mentioned inhabit the Zambianwaters. The first question one asks is how come suchisolated populations look so much alike when it isknown that they can develop many other colour pat-terns as well. The east and west coasts in the southernpart of the lake are separated by the deepest waterfound in Africa. If one is conversant with the currenttheories about the fluctuations in lake levels since theorigin of the lake, one can easily imagine a period inwhich the water dropped to a level which was lowenough to connect parts of opposite coasts. In suchsmall paleo-lakes the aforementioned species couldhave been present as a single geographical popula-tion. The rising water caused the fish to move upwardsin order to keep up with the suitable habitat (verticalmigration). This eventually resulted in the populationson either side becoming isolated. This seems to methe only plausible way to explain apparently identicalraces at both sides of the lake.

If this is indeed the only reasonable explanation thenits implications are very important. During the lowwater stand there was no water in the Zambian area.Therefore all Zambian populations must have devel-oped from individuals from the more northerly regions(horizontal migration). The fact that they almost allhave a different colour pattern leads to the conclusionthat these races must have developed in Zambian wa-ters. We may then further conclude that speciationtakes place mainly during the initial development of anew population. This, however, is at variance withDarwin’s theory that evolution is a continuous proc-ess whereby new species are slowly generated.

Combining the existence of the similar variants onboth sides of the lake with the widely accepted theoryof the fluctuations in lake level, one can only envis-age a step-wise evolution in the case of Tanganyikancichlids. This hypothesis is further supported by the“fact” that during the development of the new popu-lations in the southern part of the lake (while the wa-ter level rose) the old populations did not change at

all! These rather basic observations may lead us toconclude that when the bulk of the individuals of apopulation remain together in one large breedinggroup, the cichlids of that group will not change, prob-ably not in a million years! But when a drastic changedoes occur and new populations are founded, a newspecies could be developed almost overnight.

One could envisage the development of a new popu-lation as follows. As the rising water level opens up anew, suitable, habitat, wanderers from a nearby popu-lation could cross uninhabitable areas and settle at thenew location (horizontal migration). Of course not justone species would occur at such a new site but manydifferent species would found a new population. Whensuch new populations are founded by small numbersof individuals then there is a statistical chance that anew colour variant will develop (interbreeding). Im-portant factors in the development of a new race are,firstly, the number of wanderers founding the newpopulation. Secondly, the constitution of the localfauna and the local circumstances may greatly influ-ence the success of a particular species. The latter twofactors probably act as catalysts in speciation and thedevelopment of a new variant may depend solely onthe number of individuals at the site.

When a balance between the available species,number of individuals, and available niches has beenestablished “evolution” is suspended until a new ma-jor change occurs in this habitat.

Acknowledgement

I was greatly stimulated by the discussions I en-joyed with Martin Geerts and Mary Bailey. I sincerelythank Alain Gillot of Zaïre Cichlids for giving us theopportunity to make a survey along the Zaïrean coast.

10

The genus Xenotilapia can be divided into two groups;one group consists of maternal mouthbrooders and theother of species which form pairs during the breedingperiod and which employ the biparental mouth-brooding technique. However, this subdivision maynot always be as clear-cut as is stated here (see page18, Mark Smith’s article about X. omatipinnis).X. spiloptera belongs to the group of biparental mouth-brooders. The holotype of this species was collectedat Nkumbula Island near Mpulungu in Zambia, but itsdistribution seems to spread along other parts of Zam-bia and along the Zaïrean and Tanzanian shores.During the breeding period X. spiloptera is restricted

to the rocky habitat. This has led to the existence ofseveral geographical races. Most races have a colorlessdorsal fin with some black markings on its edge. Therace at Kigoma has a dorsal fin with tiny colored spotsbut lacking the blotchy markings found in all otherknown populations. The races inhabiting the rockyshores near Kipili in Tanzania have an attractive yel-low coloration in addition to the black markings onthe dorsal fin.

When not breeding, X. spiloptera sometimes livesin large schools over the sand. Here it forages by sift-

ing the substrate for something edible. It may inter-mingle with other species to form larger schools butmost often groups consisting only of X. spiloptera areobserved.

With the approach of the breeding season – atpresent it is not known whether there is a regular an-nual breeding season, or whether breeding is triggeredby the effect of some external stimulus on the mem-bers of the school – the school moves closer to therocky habitat and splits up into pairs. Each pair staysin a territory which is defended mainly againstconspecifics.

The pair bond is established by repeated courtship

displays by the male as well as by the female. In theartificial environment of the aquarium it seems thatthe bond between the pair is strengthened by the con-tinuous protection of the territory. If only one pair iskept in the tank there is a risk that the male and femalewill quarrel and that one of them, not always the fe-male, might end up cowering in one of the tank’s cor-ners. Even when the pair is not brooding, male andfemale stay together in a relatively small area.

The four pairs I keep in an 800 litre aquarium areeach satisfied with an area about 30 cm in diameter.

Spawning Xenotilapia spiloptera Poll & Stewart, 1975

A mouthbrooding female Xenotilapia spiloptera from Kipili, Tanzania.

Ad Konings

11

The female has just laid some eggs; the male waits behind her.

The male fertilizes the eggs when they are on the substrate.

When one pair decides to spawn, it defends an areaabout double that size. It is difficult to predict an im-minent spawning but mutual courtship increases no-ticeably, sometimes days before the actual spawning.A slight change in the color pattern occurs as well;female as well as male acquires black markings in theupper and lower part of the iris. These, together withthe pigment in the eye, form a black vertical bar acrossthe eyeball. I don’t know if this is a sign of readinessto spawn, but it is seen mostly when the pair is en-gaged in spawning or shortly before the act.

Spawning takes place inside the pair’s territory butthere is no specific site or nest where the eggs are de-posited. In fact, the deposition site may change dur-ing spawning. In the maternal mouthbrooder group ofXenotilapia, male and female circle around each otherbefore the eggs are laid, but X. spiloptera starts spawn-ing when the female suddenly deposits some eggs onthe substrate. I have never noticed a signal from themale for the female to start. He usually waits behindher, about 3 cm above the substrate, until she clearsthe site and leaves the eggs to be fertilized by him.While the male positions his vent over the eggs to fer-tilize them, the female turns around and waits untilthe male moves away. Then she picks up the eggs.

The female will not lay any eggs if the male is notbehind her. While he is chasing intruders from the ter-ritory she may remain at the site but she may also jointhe male in the territorial defence. After a short inter-ruption the female again lowers her body to thesubstrate and waits for the male to take position be-hind her before she lays a new batch of eggs.

Spawns may be as large as 40 eggs but much seemsto depend on the buccal capacity of the female. If thefemale’s mouth is getting full she has to shuffle theeggs around before she can pick up new ones. Usu-

12 The male watches the female picking up the eggs.

After fertilization the eggs are collected by the female.

ally the male has his head near the eggs when the fe-male collects them and sometimes picks up some eggshimself. Once I noticed that he picked up two eggs,which the female apparently didn’t collect quicklyenough, and swam away while chewing on them. Thenthe pair went through another cycle, and when the fe-male wanted to pick up the new batch of eggs the malespat the two eggs in front of the female’s mouth. Theywere immediately picked up by the female. Duringthe next round, the female’s mouth was probably toofull and it took so long before she had arranged theeggs already inside her mouth that the male again tooksome eggs, but this time he ate them. The female didn’tproduce more eggs after this incident.

Shortly after spawning the pair reduces the terri-tory to the usual 30 cm, but this is defended with moreenergy than usual. The first nine to twelve days thefemale broods the embryos and refrains from feeding.After this period the larvae are transferred to the male’smouth. I have never witnessed the procedure in itsentirety but it is likely that, after a set ritual, the fe-male spits all the larvae in front of the male’s mouth.The male retrieves them quickly and continues thebrooding for another ten days.

When the brooding is completed the male releasesthe fry in the territory where they are protected byboth parents. For the first few days the fry may takerefuge inside the male’s mouth but most of the timethey sit on the substrate. They sometimes wander offinto other territories where they are protected by theresident pair. The fry measure about 15 mm at the timeof release and it takes them about two years beforethey have reached the adult size of approximately10 cm.

The biparental mouthbrooding technique is not moreadvanced than the maternal mouthbrooding procedure.While mouthbrooding pairs have to devote three to

four weeks taking care of one brood, males of a ma-ternal mouthbrooder can fertilize eggs of many femalesin the same period. In theory, this should not make adifference as long as there are as many females asmales. The drawback of biparental mouthbrooding isthe fact that a territory has to be maintained by thepair. In maternal mouthbrooders only the strongestmales occupy a territory, and the mouthbrooding fe-males gather in nursery schools. Where populationdensity is high, as is the case in Lake Tanganyika, thisoffers a distinct advantage. Instead of having manyterritories spread over a large area, a species can propa-gate in a relatively small region of the biotope. Onlythe males have to worry about a territory and sincethere are fewer needed a loss during the selection proc-ess probably enhances the viability of the species.

The stimulus for the female of the biparentalmouthbrooder to transfer the larvae to the male’smouth may be hunger. During the 9 to 12 days incu-bation she does not eat at all. Females of a maternalmouthbrooder endure a longer period without food butmay eat small morsels in the second half of the incu-bation period. Such mouthbrooding females alsoexspend much less energy than those of a biparentalmouthbrooder which also have to defend territory.

In Lake Tanganyika a maternal mouthbrooder isbetter off in several respects. This could explain theobservation that breeding X. spiloptera are found atdeeper levels of the biotope than are maternallymouthbrooding Xenotilapia. Breeding pairs are usu-ally found at depths of between 15 and 40 meters .

13

The genus Tropheus

The “Murago Moorii” exists in several variants. Here a male of the “Yellow Murago” collected near Zongwe, Zaïre

Ad Konings

The species of the genus Tropheus have for many yearsbeen some of the most popular cichlids amongaquarists. There are many hobbyists who keep exclu-sively Tropheus species. Although there are only sixknown species, at least four of these show great geo-graphical variability. Fortunately most aquarists areaware of this fact and rarely house different variantsof the same species in one tank. This would lead in-evitably to bastardization.

The members of the genus Tropheus are usually wellrepresented on most rocky coasts. Except for T.duboisi, they all inhabit the upper 10 metres of thebiotope. Not only for their protection but also for theirfood, Tropheus species are restricted to areas withrocks. The feeding habits of T. duboisi have not beenextensively investigated, but those of the other spe-cies have. T. moorii, T. brichardi, T. polli and T. sp.“Black” are all grazers which shear the filamentousalgae from the rocky substrate (Yamaoka, 1983). T.annectens, which is restricted to the western, Zaïrean,shores of the lake, is closely related to and maybe evenconspecific with T. polli, and behaves similarly to thatspecies. It seems that Tropheus has a feeding relation-ship with several species of the genus Petrochromis.Members of the latter group, which are browsers comb-

ing unicellular diatoms from the algae strands, par-tially clear the biocover from sediment and make itthus more suitable for Tropheus to feed on (Takamura,1984).

In the artificial environment of the aquarium T.moorii, from the southern part of the lake, and T. sp.“Black”, form the northern, behave as strongly terri-torial cichlids. Typically the largest males will dividethe available space in the tank among themselves andleave the border areas of their territories as living spacefor females and weaker males. Territories in theaquarium measure more than 60 cm in diameter. Innature, the territoriality of these two species is not al-ways as evident as it is in a tank. Although males havea territory where spawning takes place and in whichthey forage, conspecifics are not always chased fromthe premises. Sometimes we may observe a group ofabout 20 (T. sp. “Black”) to 100 (T. moorii) individu-als grazing on the same rock. In the wild, T. brichardiand T. polli are permanently territorial and, especiallyT. brichanti, are difficult to keep in small numbers inan aquarium.

Many aquarists have great interest in maintaining abreeding group of a particular variant of one of thespecies. Not only for the continuous interactions be-

14

A male Tropheus brichardi from the Kavalla Islands, Zaïre.

tween the members of the group but also to spread theaggression of the most dominant males, it is advis-able to have as large a group as possible. If each indi-vidual is given 25 to 40 litres of water a balanced breed-ing group can be established in a suitably decoratedaquarium. Caves are not necessary as they will notprovide shelter for harassed individuals; only the domi-nant males will occupy caves. Stones should be placedin such a way that clearly separate groups are created.Each group or heap of stones, if approximately 50 cmin diameter, will be regarded as a territory by a maleand respected by the others.

Often more females are kept than males. This maywork for most species of Tropheus but not always forT. brichanti. The latter species is best kept with asmany males as females.

Before a group of Tropheus is introduced into theaquarium all of the fish must be in good condition orat least be in the same condition, and all of the fishmust be introduced at the same time. The introductionof additional specimens to the group at a later stagemust be avoided at all times! Newly introduced fishmore often than not upset the established hierarchy inthe group and frequently end up being chased by allother individuals. If the additional specimens are wildcaught fishes which have not been quarantined longenough, an early death is generally the result. Cur-rently this is the most important problem aquarists have

maintaining Tropheus.Wild caught fishes and fishes from aquaria in which

wild caught specimens are housed have many differ-ent kinds of parasites. In the wild, these parasites livein balance with the host’s immune system and arerarely present in harmful quantities. By virtue of thepresence of a low number of parasites the host con-tinuously maintains its resistance against an explosiveincrease in numbers of that parasite, although it neverbecomes totally immune. Fish in a healthy aquariumhave very few parasites and thus mostly lack this semi-immunization. When a wild caught fish is introducedto an established group, it is immediately weakenedby the repeated physical assaults it endures from allmembers of the group. This has a weakening effect onits immunity to parasites and a serious infestation isliable to result. Not only may the newly introducedfish die but the disease usually infects the nonresist-ant inhabitants as well. A single fish may thus com-pletely wipe out a long established breeding colony.Therefore, never introduce untreated wild caught fishinto a group of Tropheus.

Several parasites are specific to a group of cich-lids; they affect only those species while other spe-cies in the same aquarium do not show any sign ofdisease. Species that carry some of the Tropheus-af-fecting parasites are Goby Cichlids (Eretmodus,Spathodus and Tanganicodus), Petrochromis,

15

Tropheus sp. “Black”, a geographical variant from Kiriza, Ubwari Peninsula, Zaïre.

Simochromis, and Pseudosimochromis. There are sev-eral ways of “cleaning” wild caught fish, but all ofthem involve a quarantine period. A wild caught fish,living in balance with its parasite burden, can becleansed of them by killing the existing parasite or bypreventing re-infection where the parasites have leftthe fish in the course of their life cycle. To do this,however, one needs to know what organisms are in-volved. In Tropheus broad spectrum antibiotics andanti-parasitic solutions have had little effect in dam-ming the spread of disease. What kills most Tropheusseems to be highly infectious agents, either virus orparasites. Many aquarists have tried numerous medi-cines to cure their fish, some with success. Reasoningtells us, however, that there may be several diseasesinvolved and that thus one medicine will not auto-matically cure all fish.

As mentioned before, a fish may also be cured bypreventing the parasite from re-infecting the host. If aparasite cannot find a host it will die. The infectiousphase may live a long time but the agent infecting Tro-pheus seems to have a relatively short lifespan. Mypersonal experience and that of Hans Herrmann (pers.comm.) is that a biological trickle filter strongly in-hibits the spreading of the disease. This can be ex-plained by the fact that infectious parasites are carriedaway from the fish into the filtration system. It is ourexperience that water coming from the filter is free of

pathogens. The faster the water is recycled throughthe trickle filter the more effectively it frees it fromparasites. When the volume of water passed throughthe filter in one hour equals that of the tank’s volume,an optimum treatment has been achieved. It is notknown if faster cycles have a better or even the sameeffect. Slower cycles don’t work that well.

So, wild caught fish could be cured by placing themin a quarantine tank which is filtered by a trickle fil-ter; even when the tank is a part of a central filter sys-tem! It is important to leave the fish for at least twomonths on their own in that quarantine tank and havea separate hand-net for that tank only. When directwater-to-water contact with other tanks is avoided theresult is healthy fish.

References

TAKAMURA, K. (1984) Interspecific relationships ofaufwuchs-eating fishes in Lake Tanganyika. Env.Biol. Fish. 10 (4); pp 225-241.

YAMAOKA, K. (1983) Feeding behaviour and dentalmorphology of algae scraping cichlids (Pisces:Teleostei) in Lake Tanganyika. Afr. Study Monogr.4; pp 77-89.

16

Since its description 35 years ago, this cichlid has beenknown to the scientific world but has never been in-troduced into the aquaristic hobby. Its rather unexcitingcoloration could be one of the factors that postponedits entry into the hobby; on the other hand the firstfew specimens were only recently collected alive bysome German aquarists. Until then its only knownoccurrence was located on the northwestern coast ofthe lake, along the Zaïrean coastline. Now a secondlocation has been found on the Burundi coast in thenortheastern part, just north of Nyanza.

The fact that I discovered Simochromis marginatusin the rocky habitat at this location made me a littleuneasy. This part of the lake, which is easily accessi-ble, has been thoroughly investigated by the late PierreBrichard. This author has denied, in his book “Cich-lids and all the other fishes of Lake Tanganyika”(1989), the existence of S. marginatus in this part ofthe lake. My unease is increased by the fact that S.margaretae – this species has not yet been exportedalive – is recorded from a location not far from theplace where I found S. marginatus, i.e. Kigoma Bay.S. margaretae is also characterized by a black mar-ginal band in the dorsal fin and has a very similar shapeto S. marginatus.

Not until I had examined the specimens and com-

Simochromis marginatus Poll, 1956Hans-Joachim Herrmann

A wildcaught male Simochromis marginatus from Nyanza, Burundi. Photo by Hans-Joachim Herrmann.

pared them with the original descriptions of the twospecies could I conclude that I had collected S.marginatus. The only difference is the relatively deepercaudal peduncle of S. margaretae.

Like all other species of the genus S. marginatusspawns easily in captivity. The male constructs a shal-low spawning-dish in the sand and leads ripe femaleswith undulating and vibrating movements into his nest.When both partners are ready to spawn they circle thespawning-site. The eggs are fertilized inside the fe-male’s mouth. When all eggs are laid the female re-treats from the male’s territory and cares for the eggsduring the following three weeks. After this period allegg yolk has been absorbed and the fry are about 1 cmlong. Only when danger threatens are they taken backinside the female’s mouth.

At a size of about 3 cm juvenile S. marginatus al-ready show intraspecific aggression, and this is knownto occur among the other species of the genus as well.Such aggression can easily be reduced by accompa-nying them with other non-related species like Troph-eus moorii or a species of the genus Petrochromis.Males can reach a maximum size of approximately 11cm while females, which do not show a distinct blackmarginal band in the dorsal fin, remain about 2 cmsmaller.

17

The genus Simochromis consists of small to mediumsized mouthbrooding cichlids which inhabit the rockyshallows or the intermediate biotopes in Lake Tan-ganyika.

One of the smaller species is Simochromis babaultiwhich has a maximum length of about 10 cm andwhich is fairly well known among aquarists. Becauseof its rather aggressive — intraspecific — behaviour,and its lack of a sparkling coloration, S. babaulti ismaintained by only a few specialists.

In 1989 I had the opportunity to bring back alive anew variant of this species which had not been knownbefore. Males of this variant show yellow instead ofbeige vertical bars. Moreover there are a number ofbright red spots sprinkled over the upper part of thebody. The latter feature reminds us of S. pleurospilus(see photo) and possibly these two species have a muchcloser relationship than has been thought previously.Females of this variant, which have a maximum sizeof approximately 8 cm, have a very attractive colourpattern consisting of a large red blotch covering thehead, ventral fins and part of the body. The intensityof the red pigment varies according the emotional stateof the fish. I found this variant in a shallow intermedi-ate habitat a few miles north of Nyanza, Burundi.

Simochromis babaulti Pellegrin, 1927Hans-Joachim Herrmann

An adult male Simochromis pleurospilus from Zambia.

Simochromis babaulti; a submissive male is visible in the background. Photos by Hans-Joachim Herrmann.

A female Simochromis babaulti from Nyanza, Burundi.

18

Xenotilapia ornatipinnis Boulenger, 1901Mark Smith

A courting male Xenotilapia ornatipinnis. Photos by Mark Smith

A brooding female Xenotilapia ornatipinnis.

Xenotilapia ornatipinnis has recently been importedunder the name of “Pearly Xenotilapia”. The fishescurrently in the hobby are collected from the northernpart of the lake and were exported by Fishes ofBurundi.X. ornatipinnis is known to occur down to depths of150 metres (500 feet). However, it also occurs in muchshallower waters, a fact which has enabled its collec-tion. As with several species of Tanganyikan cichlids,X. omatipinnis seems to display a wide depth rangepossibly related to the daily vertical migration of plank-ton on which it feeds. X. omatipinnis displays a“pearlescent” body coloration so very characteristicof deep living, sand-dwelling cichlids from the lake.

Its large, flattened eyes are also marvelous adaptationsfor living in poorly lit environments. The maximumrecorded size is 13 cm total length.

In the aquarium, X. omatipinnis readily adapts tothe regular aquarium fish food. It behaves very peace-fully; never did I observe any torn fins or any indi-viduals cowering in the corners of the aquarium.

X. omatipinnis has been bred in the aquarium andin most instances the female broods her embryos tofull term on her own. Ron Sousy (pers. comm.) found,however, that the female would sometimes give theseveral days old embryos to the male who would thenincubate the developing embryos to full term. On oneoccasion a carrying male gave the developing embryosto yet another male! This foster father held for a littlemore than one day before disposing of the brood. Toknow whether or not this is a naturally occurring eventis difficult to ascertain but it seems unlikely that a spe-cies which does not seem to form pairs and whosemales spawn with several females in a few days, wouldnormally behave in this manner.

X. ornatipinnis is thus a very interesting Tangan-yikan cichlid which may provide us with even moresurprises. Its refined attractiveness and peacefulbehavior should help to establish its popularity amongaquarists in the years to come.

19

Ctenochromis benthicola (Matthes, 1962)Mark Smith

A female Ctenochromis benthicola. This specimen was imported as “Orange Compressiceps”. Photographer unknown.

One of the most unusual and rare cichlids presentlyknown from Lake Tanganyika is Ctenochromisbenthicola. The specimen in the accompanying pho-tograph is a female from the northern end of the lake.This individual somehow made its way into the U.S.A.in the early 70’s under the name of “OrangeCompressiceps”. Whether or not more of these cich-lids were ever brought out of the lake for the aquariumhobby is uncertain.

Before Poll’s revision of the Tanganyikan cichlids(1986) this species was included in the genusHaplochromis. Now it is assigned to Ctenochromis,which seems to be an odd genus since representativesof it are found well outside of Lake Tanganyika andseem to be of a different lineage.

Males of this species are a dark brown with a darkblue overlay, while females are entirely orange withjust a few black specks on the sides of the body. Thispeculiar color scheme is unique among Tanganyikancichlids. The brilliant orange color of the female wouldseem to make it more conspicuous.

The intense competition in the shallow waters ofLake Tanganyika explain why this fish inhabits onlythe deeper waters. Its scarcity in the lake may also bedue to the greater success of other cichlids causing C.benthicola’s numbers to dwindle, indicating that it is

on its way to a natural extinction in the years to fol-low. On the other hand some species naturally havesmall numbers because that keeps them in balance withtheir biotope.

C. benthicola is also endowed with enlarged sen-sory pits in the head, a feature shared with membersof the genus Trematocara, which also inhabit greatdepths. These enlarged pits probably enable these cich-lids to find their food in the relative absence of light.

C. benthicola was successfully bred by theBrichards in the mid 70’s. Curiously details of its re-production were absent from their report, but sinceonly a very few juveniles were produced it may bethat C. benthicola is a mouthbrooder like all otherhaplochromines. The juveniles were brown at a lengthof about 5 cm and thus looked like their father. Somequestions must arise from this bit of information. Didthe spawn contain only males or do females start outlooking like the brown males and then change coloras they mature. Another possibility is that the orangefemales are a kind of color morph, like the O and OBcolor morphs among Malawian cichlids. (Accordingto Mireille Schreyen from Fishes of Burundi, brownfemales occur as well making it likely that the orangeindividuals are indeed a polymorph. Ed.).

20

Neolamprologus petricola (Poll, 1949)Ad Konings

A wild caught male Neolamprologus petricola.

The name “Lamprologus Petricola” has been aroundin the hobby for a long time. It was given to the yel-low morphs of Neolamprologus mustax and seems stillto be in use. Interestingly, Neolamprologus petricolaresembles N. moorii and N. modestus more than N.mustax. The name petricola means “living amongrocks” and alludes to the assumption that this cichlidis restricted to rocky habitats. N. petricola inhabitsthe intermediate biotopes along the southwestern coastof the lake. It has never been observed in Zambia. Al-most all specimens exported were collected south ofMoba in Zaïre, but it is found as far north as Cap Tembwe.

It is very difficult to formulate a distinction betweenN. petricola and N. modestus, especially when semi-adult. Both species have an elongated shape, but, ingeneral, N. modestus has a shallower body. Both spe-cies have a dark brown coloration but under subopti-mal conditions both species show a much lighter pat-tern. N. modems, in such circumstances, has a muchyellower tinge than the beige-greyish colored N.petricola. Submissive coloration of the latter speciesshows grey, broad vertical bars on an almost whitebackground. I have never seen the forms of N.modestus from areas other than the deep south of thelake and these types are characterized by bright yel-low pectoral fins. The pectorals of N. petricola are

colorless or greyish but never yellow. Large specimensof N. petricola show a cranial gibbosity which hasnever been observed in N. modestus. The latter spe-cies has thicker lips too, but the teeth on the pharyn-geal jaws are similar and indicate a diet of hard inver-tebrates for both species. The two species may differin habitat preference as N. petricola is also found atdeeper levels whereas N. modestus seems to be re-stricted to the shallow regions of the sediment-richintermediate biotope.

The first observations in the aquarium indicate thatN. petricola is a territorial cichlid which is difficult tohouse with conspecifics in a small tank (less than 200litres). The sexes are easily recognized by examiningthe vents and thus a single pair can be selected andplaced in a large tank. It is the female who starts dig-ging a nest under a rock and she tries to attract themale to her nest.

N. petricola grows to a size of about 13 cm and itseems to attain the deeper body and the gibbosity onthe head only at adult size. As in N. moorii the femaleis slightly larger than the male and she is able to pro-duce large spawns. A spawn from a relatively youngfemale numbered more than 300 fry and it seems thatfully grown females are able to produce more than 500eggs per spawn. The juveniles have a light grey color.

21

Neolamprologus leloupi (Poll, 1956)Ad Konings

Neolamprologus leloupi at Kanoni, Zaïre.

In the first volume of the “Cichlids Yearbook” I havewritten an entry about a Neolamprologus with a yel-low dorsal fin which I named N. leloupi. At that timeI had not observed this species in its natural environ-ment and agreed with Brichard (”Cichlids and all theother fish of Lake Tanganyika”, 1990: 537), who hadsurveyed the entire Zaïrean coast, that N. caudo-punctatus is a synonym of N. leloupi.

In September 1991, Martin Geerts and I made ashort visit to the southern Zaïrean part of the lake andalso dived near a village named Kapampa. In this areathe distributions of N. leloupi and of N. caudopunctatusoverlap. This is also the only location where N.caudopunctatus has a bright yellow dorsal. This mayhave been the sole reason why these two species donot hybridize. N. leloupi is distributed north fromKapampa up to Cape Tembwe whereas N. caudo-punctatus is found south from this region (includingthe entire Zambian shoreline of the lake).

N. leloupi occurs in several population which dif-fer slightly in the markings in the fins. There is nodifference between its behaviour and that of N.caudopunctatus, and at Kapampa groups of both spe-cies are found side by side in the intermediate habitat.They occur from the shallows to a depth of about 25 m.

Neolamprologus leloupi at Kapampa.

Neolamprologus caudopunctatus at Kapampa.

22

Neolamprologus sp. “Ubwari Buescheri”Mark Smith

One of the first exported Neolamprologus sp. “Ubwari Buescheri”. Photo by Mark Smith.

The existence of this undescribed Neolampmlogus sp.“Ubwari Buescheri” was first made known photo-graphically to the aquarium world in Axelrod’sAquarium Atlas (1985; first ed. p 511, plate # 349)where it was wrongly identified as Lamprologusbuescheri. According to Pierre Brichard (1990), hisson Thierry discovered this lamprologine at the UbwariPeninsula in February of 1984. Konings (1988) re-ports that a specimen of this species was found to-gether with a collection of other cichlids from Burundi,which could indicate that it is also found in this partof the lake. (It has not yet been observed in Burundiwaters. See next page for the discussion of a similarspecies from the Kavala Islands. Ed.)

There are a number of differences between this spe-cies and N. buescheri. The “Ubwari Buescheri” hassmaller, more numerous scales than N. buescheri fromthe southern part of the lake. The horizontal stripingin N. sp. “Ubwari Buescheri” is much narrower – itcan at times even fade completely in some aquariumspecimens – than that of N. buescheri, The basic pat-tern of N. buescheri consists of two horizontal stripesand several vertical stripes. Depending on the loca-tion either the vertical striping (Gombi population) orthe horizontal striping (Kachese population) is pre-dominant or else a combination of both patterns is

present (Chituta Bay population). N. sp. “UbwariBuescheri” does not show such geographical varia-tion. In fact the upper horizontal stripe of this speciesis at a noticeably different angle than in N. buescheri.Furthermore N. sp. “Ubwari Buescheri” shows cra-nial gibbosity at a small size, 6-7 cm TL, suggestingthat 6-7 cm constitutes the adult size for this fish. Thiscranial gibbosity seems to be lacking in N. buescherifor the most part. Lastly, from superficial observations,the trailing ends of the unpaired fins of N. sp. “UbwariBuescheri” tend to be finer or threadlike in contrast tothe thicker fin-tips seen in N. buescheri. All these ob-servations plus further scientific analysis of thislamprologine will undoubtedly show that these twocichlids are not conspecific.

N. sp. “Ubwari Buescheri” was found at a depth rangeof 15 to 25 metres, a range that basically matches that ofN. buescheri. It has also proven itself a territorialaquarium resident; like N. buescheri it forms pairs anddefends a small section of the aquarium as a territory.The five specimens, two females and three males, wekeep in our tanks have yet to reproduce, but it is prob-ably safe to say that they spawn in the same fashion assimilar-sized, rock-dwelling lamprologines do. The finelyelongated fins make N. sp. “Ubwari Buescheri” a grace-ful addition to the spectrum of Tanganyikan cichlids.

23

Neolamprologus sp. “Kavalla”Ad Konings

A male Neolamprologus sp. “Kavalla” photographed at Milima Island at a depth of about 15 m.

Neolamprologus sp. “Kavalla” closely resembles N.sp. “Ubwari Buescheri” (see previous page) in shapeand in some of its pigmentation pattern but grows to amuch larger mature size. The maximum size of N. sp.“Kavalla” is estimated at 12 cm (no specimens werecollected for closer examination).

I have observed this attractive cichlid at Milima Is-land, one of the islands in the Kavalla archipelago northof Kalemie on the Zaïrean shores of the lake. Itsbiotope consists of large, rounded boulders whichshowed at the time of my visit (September 1991) justa thin layer of biocover. The coast around the islandshelves gradually to a depth of about 30 m where thehabitat becomes predominantly sandy. N. sp.“Kavalla” is seen from 10 m on to the edge of therocky part at about 30 m. It normally occurs solitaryand is not common. One would expect that, like N.buescheri, males at least would have territories, butterritorial individuals were not seen. It seems that N.sp. “Kavalla” roams through the biotope in search offood which presumably consists of invertebrates. Itwas frequently seen swimming in and out of caves inwhich it may find most of its food. In this respect itbehaves more like N. leleupi than like N. buescheri.

Only the larger individuals, presumably males, showa grey-blue colour on head and body. Smaller speci-

mens, probably females and sub-adults, rarely displaythe grey-blue colour. These individuals show a basicpattern of two horizontal lines in which they resembleN. sp. “Ubwari Buescheri”.

All adult specimens have a dark grey-brown patchon the lower part of the head, which is most intenselycoloured near the chin where it abruptly disappears,leaving a pure white chin. Although the observationthat adult size and coloration differ from those of N.sp. “Ubwari Buescheri”, it is still possible that we aredealing with two populations of one species. Not un-til the rocky regions between Ubwari and Kavalla havebeen investigated and no intermediate Lamprologineshave been found may we conclude that these two popu-lations belong to different species. The basic pigmen-tation pattern, i.e. the two horizontal lines on the body,is seen in a small number of cichlids belonging toNeolamprologus as well as Lampmlogus. It seems tobe a characteristic of cichlids which are rare and usu-ally live at deep levels. Besides the two species men-tioned here there is also L. sp. “Zambia” (see photo in“Tanganyika Cichlids” by Konings, 1988: 108) andfrightened specimens of N. sp. “Cygnus” (see photoin “Cichlids Yearbook”, vel. 1, 1991: 11) having sucha colour pattern.

24

Cyphotilapia frontosa (Boulenger, 1908)Ad Konings

The race of Cyphotilapia frontosa at Kavalla, Zaïre (photo above) resembles that of Burundi.

Plecodus straeleni photographed in Burundi waters.

The first live specimens of Cyphotilapia frontosa wereexported by Fishes of Burundi in the late fifties. Fromthen on it has been a mainstay in the aquarium hobbyand become widespread among aquarists. C. frontosahas a lake-wide distribution and occurs in several distin-guishable races. Initially only the populations of Burundiwere exploited. Then the Kigoma race was introducedto the hobby. The latter variant shows one vertical bandmore on the body than the Burundi race and has thusbecome known as the Seven-Band Frontosa. In view ofthe other variants known to date it is better to refer to itas C. frontosa (Kigoma).

It is characterized mainly by the dark blue colour onthe cheeks and the variable yellow coloration in the dor-sal fin. The third variant exported from the lake comesfrom Zambian waters. It is characterized by a brighterblue coloration (in comparison with the northern races)and an interorbital band. In the late eighties a very densepopulation of C. frontosa was discovered near the Kavallaarchipelago in Zaïre. This variant resembles the Burundirace but occurs in rather shallow water (up to 5 metres).

In 1990 Alain Gillot (Zaïre Cichlids) discovered a newand exciting variant of the frontosa. This population ischaracterized by the intense blue colour on all fins andthe mother of pearl on the hump and upper part of thebody in large specimens. Juveniles also show the attrac-tive blue coloration seen in adults. There is, however,one drawback; it occurs at deep to very deep levels. Theshallowest point I have seen this blue frontosa was 25metres. Most specimens are collected at a depth between40 and 60 metres and require three days to decompress.The blue variant is distributed on the rocky shores be-tween M’Toto and Kapampa on the Zaïrean coast.

An interesting phenomenon is the occurrence of ascale-eating cichlid, Plecodus straeleni, which has anidentical coloration to that of C. frontosa. It was re-ported (Brichard, 1978, 1989; Krüter, pers. comm.)that P. straeleni uses its coloration as a camouflage

25

The blue race of Cyphotilapia frontosa is distributed along the southern part of the Zaïrean coast.

Plecodus straeleni at Kapampa, Zaïre.

which enables it to mix with groups of C. frontosaand stealthily scrape some scales from the flanks ofthe surprised frontosas. I have never observed P.straeleni attacking C. frontosa although I have seenthem among their schools. However, I have observedP. straeleni attacking other cichlids, like Cyathophar-ynx furcifer or even the much smaller Neolampmlogusbrichardi.

It has also been said that it may mimic not only C.frontosa but also Neolamprologus sexfasciatus andN. tretocephalus. There are several reasons why thisseems unlikely. First of all P. straeleni resembles C.frontosa to such an extent that it is easily mistakenfor its model, even at close quarters. It goes to greatlengths to copy C. frontosa, so much so that the popu-lation at Kapampa has a much bluer coloration (likeC. frontosa) than that at Rutunga, Burundi (see pho-tographs). Secondly, in Burundi waters there are noN. sexfasciatus or N. tretocephalus, which would ofcourse not exclude the theory that P. straeleni’s col-our pattern could have developed in mimicry of thosespecies, but it would make the extreme resemblanceto C. frontosa at locations where these two speciesare found inexplicable. Thirdly we could argue thatother scale-eaters, especially the abundantly presentPerissodus microlepis, do not mimic their prey. If we

combine this with the observation that P. straeleni at-tacks other cichlids than C. frontosa then it seemsunlikely that P. straeleni needs its coloration to beable to exist predominantly on scales of C. frontosa.

C. frontosa roams about in the rocky biotope but ithas never been seen hunting prey. Other species in-habiting the same biotope probably know its peacefulmanners and thus give P. straeleni, dressed in a“sheepskin”, a good opportunity. Feigning to be agood-natured C. frontosa, the scale-eater comfortablycloses in on its prey and before the victim recognizesthe wolf it has lost some of its scales.

26

Cyprichromis sp. “Leptosoma Jumbo”Ad Konings

The yellow morph of Cyprichromis sp. “Leptosoma Jumbo” at Kitumba, Zaïre.

The genus Cyprichromis consists of several specieswhich belong to the most colourful cichlids of LakeTanganyika. Representatives of the genus are foundat most rocky regions around the lake but none of thespecies has a lake-wide distribution. Only two spe-cies are scientifically described but at least two othersare known to occur. At some locations in Zambia (e.g.Mpulungu; René Krüter, pers. comm.) and Tanzania(e.g. Malasa Island; Walter Dieckhoff, pers. comm.)two species of Cyprichromis are found sympatrically,but at most other locations it seems that only one spe-cies of this genus occurs. However, they are usuallyfound sympatrically with Paracyprichromis brieni andP. nigripinnis.

It is generally known that in a single population ofCyprichromis males with a noticeably different col-our pattern, especially in the tail, occur in the sameschool. I have suggested before (1988) that the blue-tailed individuals could in theory be a genetically dif-ferent species from the yellow-tailed specimens. Theirbehavioural preference for feeding from planktonabove rocky substrates may have brought these twotheoretical species together as it brought Para-cyprichromis, which has a similar feeding behaviourbut is from a different ancestry, together withCyprichromis.

The observation that a female may give birth toyellow as well as blue tailed males in captivity doesnot prove the natural cause of things. Such a female,e.g. originally a yellow-tailed “species”, may havespawned with a blue-tailed male in the confines of anaquarium. If the colour of the tail constitutes the maincriterion for mate-recognition, such a female wouldnever have done so in the wild. Until unambiguousexperiments have been performed revealing the in-heritance of the colour of the tail, the possibility oftwo sympatric species cannot be excluded.

Sometimes, when such “hybridization” experimentsseem to have occurred in the wild, we may deducefalse clues regarding the specific status of the differ-ent sympatric species. This may be the case with thepopulation of Cyprichromis at Kitumba, Zaïre. In thispopulation not only the colour of the tail differs butalso the colour on the body. In fact there are three dif-ferent morphs (or species?): there are completely yel-low coloured males and blue coloured males with yel-low and blue tails.

I could observe this school for only one hour butduring this short period I gained the impression thatthe complete yellow individuals and the blue individu-als represent two different species. As is usual amongCyprichromis, males defend their three-dimensional

27

The dorsal fin of the male of the blue morph (here with yellow tail) becomes almost white when he courts.

Notice a small aberration in this male’s tailfin.

A male of the yellow morph with a broad blue segment.

territory in the open water and relate the boundariesto the distance from neighbouring males. Yellow andblue males were found mixed and among them largenumbers of females, which all looked identical, triedto find a partner to spawn. Both types of males some-times courted the same female but their territorial ag-gression was always directed towards a male of thesame colour; i.e. yellow males would chase only otheryellow males while blue males (with blue or yellowtails) chased only other blue males. Sometimes a chasefollowed a route through the territory of a differentlycoloured male!

After a while I noticed a few rare individuals whichlooked like a cross between the two morphs. AlthoughI do not know what a cross between these two morphswould look like, the colour patterns these individualsshowed did not indicate that a cross would yield simi-lar looking offspring. The colour aberration I noticedand photographed (see accompanying pictures) weremainly a type of local mutation rather than a recurrentpattern of a segregation of the parental colors. So therewas a mainly yellow individual with a blue blotch onits body or vice versa, a blue male with a yellow spot.Although it cannot be denied that the yellow malesare distinctly different from the blue ones, it is stillnot clear whether females appreciate these differences

28

The completely blue morph of Cyprichromis sp. “Leptosoma Jumba” from Kitumba.

as well. The fact that I could find three aberrant indi-viduals among a few hundred males may indicate thatthey are not products of an extremely rare event.

These observations can be explained in several dif-ferent ways of which true hybridization of twosympatric but distinct species is just one. The yellowand blue males could represent different morphs ofone species and a female would spawn with all threeof them. The territorial colour of the males could beregulated by one or very few genes which could actlike a switch in a juvenile phase of the male: normallyit would switch on either blue or yellow but in someindividuals and at certain regions of the body the switchcould have been reversed. This could occur as aphenotypic event, independently from genetic factors.

If the males of this population are to be regarded asmorphs of one species then they certainly present aremarkable case of polymorphism (see photographsfor how different the two morphs are). If the twomorphs represent two species then the definition of anatural species (specific recognition under natural cir-cumstances) seems to be in need of a revision becausefemales apparently can mistake a male of the otherspecies as their true mate. Mary Bailey (pers. comm.),however, suggests that misfertilisation could easilyoccur in a mêlée, and that the odd males I saw were

probably an insignificant part of the total population.If such contrastingly coloured males can belong to

one species then how can we explain the existence oftwo species of Cyprichromis at Malasa Island andMpulungu? The colour difference between the malesof these two species is much less dramatic than in theKitumba population. Maybe the latter population canteach us a lesson on speciation. It may turn out thatcoloration is not the most important criterion in spe-cies recognition but that species specific odours com-bined with behavioural and morphological charactersplay a major role instead.

A possible advantage for Cyprichromis having dif-ferently coloured males which do not visually recog-nize each other as belonging to the same species, isthat their territories can be closer together without elic-iting constant territorial fights of similar looking,neighbouring males. Indeed I frequently observedyellow and blue territorial males almost perfectly al-ternating in the water column, For the time being Iregard the Kitumba population as belonging to one,polymorphic species, namely Cyprichromis sp.“Leptosoma Jumbo”.

29

Paracyprichromis brieni (Poll, 1948)Ad Konings

A wildcaught male Paracyprichromis brieni collected at Kitumba, Zaïre. Note the spots in the trailing part of the dorsal fin.

A female Paracyprichromis brieni (Velifer).

When the first specimens of this pretty little cichlid –with the trade name “Cyprichromis Velifer” – arrivedin western aquarists’ tanks it was immediately thoughtthat the individuals with the attractive black and whitedorsal fins were males and the ones with the plain finswere females. Of course, after acclimatization and withobservations made under less stressful conditions itbecame apparent that it is the female which is adornedwith the black and white dorsal fin. Due to the lack ofyellow coloration on the tips of the ventral fins thisspecies is placed in the genus Paracyprichromis ratherthan Cyprichromis.

It is a geographical variant of P. brieni which isdistributed along the Zaïrean coast between Kitumbaand Moba. It has a rather wide depth-range since indi-viduals were observed between 5 and 35 metres.

Since its introduction it has been bred in captivity,but reports concerning a spawning are not yet avail-able. Courting behavior differs slightly in some pointsfrom that of P. nigripinnis. Males occupy territories,like P, nigripinnis males, but these are not necessarilyalongside the vertical face of a rock or at the side ofthe tank. Females stay in the open water as well andform schools. Male P. nigripinnis court females bydisplaying themselves with all fins erect, followed byleading the female to the centre of its territory whileswimming in an undulating fashion.

The courting behavior of P. brieni (Velifer) differsin that some males do not lead the females whilevigourously quivering their bodies. The male startscourting with a display of all its fins accompanied bya slight quivering and sometimes sideways jerking ofthe body. Meanwhile he inches his way towards thefemale and, at close range, gives her a sudden blowwith his tail. Males P. brieni (Velifer) are character-ized by a black spot in the middle of the tailfin. As faras is known this is the only geographical variant of P.brieni with such a feature.

30

A male Cynotilapia afra in the rocky habitat around Chitandi Island; note its teeth!. Photos by A. Spreinat.

MALAWIANCICHLIDS

The genus Cynotilapia Regan 1922

Dr. Andreas Spreinat

The main typical feature of the species of the genusCynotilapia is unicuspid teeth, spaced relatively farapart and resembling the teeth of a predator. Most othermorphological characteristics resemble those of themembers of the Pseudotropheus zebra complex to suchan extent that without knowledge of the tooth struc-ture, they can easily be confussd with these mbuna. Innature another feature can be observed that furtherdifferentiates Cynotilapia from Pseudotropheus: incontrast to most other mbuna, Cynotilapia, except forterritorial males, forages in the water column severalfeet above the rocky substrate. Their food consists of

plankton rather than of Aufwuchs. The unicuspid,widely spaced, teeth are better suited to the capture ofsmall planktonic invertebrates (zooplankton) than tocombing the algae of the biocover.

The type species, C. afra, was described in 1893by the ichthyologist Albert Günther, who worked atthe British Museum (Natural History). In 1976, thesecond species in this genus, C. axelrodi, was describedby Burgess. Tony Ribbink and his co-workers, a SouthAfrican team of ichthyologists, concluded, after theinvestigation of many rocky habitats along the shoresof Lake Malawi, that a further 8 species of Cynotilapia

31

A male with a black dorsal. Is this a geographical variant of Cynotilapia afra or a different species?

exist (Ribbink et al., 1983). These species were notformally described but given provisional names.

Probably even more species exist. Some time agospecimens from a population of Cynotilapia wereimported under the trade name “Jaro Afra”. These cich-lids, collected at Jalo Reef near Nkhota Kota, belongto a scientifically undescribed species provisionallycalled C. sp. “Jalo”.

Among aquarists C. afra is the best known species.I am not aware when it was exported for the first time,but it must have been in the late sixties or early seven-ties. C. afra is found in the northern half of the lake.Likoma, Chizumulu and the rocky shores betweenChirombo Point and Ngara on the West coast areknown to harbour this mbuna.

The variation in the colour pattern of C. afra is note-worthy. On the one hand populations show geographi-cal variation in male breeding coloration, and on theother hand variations exist within a single population.These variations are not just subtle changes in the col-our pattern but include conspicuous differences in thenuptial coloration of males, especially the colorationof the dorsal fin. Ribbink and his coworkers investi-gated the differences in the coloration of the dorsalfins of males in the populations inhabiting the north-ern shores of Likoma Island. They found that the

number of males with black dorsal fins decreased to-wards the south, whereas males with white or yellowdorsal fins became more abundant there.

Among aquarists it was previously common prac-tice to differentiate between Cynotilapia andPseudotropheus by looking at the vertical bars on thebody. These continued into the dorsal fin in Cynotilapiaand not in Pseudotropheus. Owing to the increasedknowledge of species and geographical variants thiscriterion has been invalidated, as some species ofCynotilapia are known to have entirely white dorsalfins while some members of the Pseudotropheus ze-bra complex have vertical barring extending into thisfin. Given the considerable variation known to occurwithin a population of C. afra, it is difficult to assesswhether in other populations we are dealing with ageographical variant or with a different species. Com-plicated and elaborate investigations are necessary inorder to make a final pronouncement.

By comparison the recognition of C. sp. “Mbamba”as a valid species is relatively simple. C. sp. “Mbamba”lives sympatrically with C. afra at most locations al-beit at a generally deeper level. Also C. sp. “Mbamba”shows a considerable variation in male breeding col-oration, geographically as well as within a single popu-lation. Some individuals are almost completely black,

32

A male Cynotilapia sp. “Mbamba” at Chitande Island.

i.e. the black bars are so broad that they cover most ofthe flanks. Other specimens, however, have a colourpattern not unlike that of Pseudotropheus zebra. Somemales have a yellow, and others a light blue, blaze ontop of the head. This part of the head is also the sim-plest feature by which one can discriminate betweenC. afra and C. sp. “Mbamba”. C. afra has a black in-terorbital bar (from eye to eye) and a bar across theupper part of the head. This second bar, which runsbetween the upper parts of the gillcovers, is lacking inC. sp. “Mbamba”. Sometimes a faint trace of the baris visible but the background colour, yellow or lightblue, is clearly visible giving the fish the appearanceof having a coloured cap.

Unfortunately C. sp. “Mbamba” was exported onlysporadically although it would make a very attractiveaddition to the mainstays of the hobby. C. axelrodi,however, is a frequently exported member of the ge-nus. It is usually exported under the trade name“Kingsizei”. The first specimens were exported byPeter Davies in 1973. For a long time the identity ofC. axelrodi was uncertain, probably because Burgess,who described the species in 1976, gave as type local-ity the southern part of the lake (Burgess, 1976: 41).This was the location from which the exporter oper-ated but not where this ’species was subsequently

found. Ribbink and his coworkers (1983) and severaldiving aquarists found C. axelrodi to be restricted tothe western coast between Chirombo Point and Li-on’s Cove. Another species, which belongs toPseudotropheus because of its bicuspid teeth, collectedat Likoma Island and also named “Kingsizei”, furtherconfused the identity of C. axelrodi.

Not only does C. axelrodi differ from the other spe-cies of the genus in its almost nondescript, light bluecoloration, but it also has an elongated shape. Thepopulation which I was able to observe north of NkhataBay inhabited shallow water over a mixed bottom at adepth of 4-6 metres. Males behaved territorially anddefended areas with a diameter of about one meter.The only other Cynotilapia which is elongated like C.axelrodi, is the scientifically undescribed C. sp. “Lion”.This cichlid, with its trade name “Lions Cove Afra”,is, as its name indicates, collected near Lion’s Coveon the northwestern shore. C. sp. “Lion” also prefersa mixed sandy-rocky bottom in shallow water. It israrely found in waters deeper than 20 metres. Malesexcavate small caves into which females are led inorder to spawn. Like many other mbuna, C. sp. “Lion”tends to lose its elongated body-shape when fed copi-ously. Such specimens are often as deep-bodied as wildC. afra.

33

Cynotilapia sp. “Chinyankwazi” at Chinyankwazi Island.

A male Cynotilapia sp. “Lion” at Lions Cove.

C. axelrodi 2 km north of Nkhata Bay.

The distribution of C. sp. “Chinyankwazi” is lim-ited to the two small islets, Chinyankwazi andChinyamwezi, in the southeastern arm of the lake.(Recently a population has been discovered atChinyamwezi Reef, Ed.). These two islets belong tothe National Park and are forbidden grounds for thecollection of ornamental fish. The coloration of C. sp.“Chinyankwazi” is almost identical to that ofPseudotropheus zebra. Interestingly, P. zebra is notpresent at these two islets. It would be worth the ef-fort to investigate how the niche of C. sp. “Chin-yankwazi” compares with that of P. zebra. It is possi-ble that C. sp. ’Chinyankwazi” has been exported inearlier times, when collection at these two islands wasstill allowed. Maybe some specimens of this mbunaare still around in the hobby, possibly under the nameP. zebra or as an unidentified species.

The other species, for which there is no space forphotographs, will be discussed for completeness.Ribbink and his coworkers identified a rare species atNdumbi Rocks, Likoma Island with the name C. sp.“Ndumbi”. It lives in secluded caves in shallow wa-ter. Its coloration is almost entirely black.

C. sp. “Maleri” is known only from the Maleri Is-lands (Ribbink et al., 1983). It has a dark blue bodywith black vertical bars. The Mbenji Islands harbourtwo species of this genus, C. sp. “Yellow Dorsal” andC. sp. “Black Dorsal”. The main characteristics of themale breeding coloration are used in their unofficialnames. C. sp. “Jalo” was mentioned earlier in this ar-ticle.

The maintenance of Cynotilapia is not much dif-ferent from that of other mbuna. Besides a slightlyalkaline pH-value of the water and a modest feedingregime, the size of the tank and its decoration is animportant factor. Under the right conditions and witha restricted but varied diet Cynotilapia behaves as alively and robust mbuna. As with many other mbuna,territorial Cynotilapia can be aggressive towardsconspecifics as well as other species.

Most of the imported species can be bred in captiv-ity. However, little has been reported about the inher-itance of the different colour patterns in C. afra andC. sp. “Mbamba”, the two species with the variablecoloration. There is thus an opportunity, especially foraquarists, to discover some interesting facts about thesefascinating mbuna.

References

BURGESS, W. E.. (1976) Studies on the familyCichlidae: 7. Cynotilapia axelrodi, a new species

of Mbuna from Lake Malawi (Pisces: Cichlidae).Trop. Fish Hobbyist 24, Oct. pp 37-44.

GÜNTHER, A. (1893) Second report on the reptiles,batrachians and fishes transmitted by Mr. H. H.Johnson, C. B., from British Central Africa. Proc.zool, Soc. Lond. pp 616-628.

RIBBINK, AJ., B.A. MARSH, A.C. MARSH, A.C. RIBBINK,and B.J. SHARP (1983) A preliminary survey of thecichlid fishes of rocky habitats in Lake Malawi. S.Afr. J. Zool 18 (3), pp 149-310.

34

The race of Protomelas taeniolatus at Chinyankwazi Island is better known as “Haplochromis Fire Blue”.

The Protomelas taeniolatus-complexAd Konings

Protomelas taenioIatus is one of the most frequentlyexported non-mbuna from Lake Malawi. In the aquaristictrade it is better known as Haplochromis Steveni. Dur-ing the lake’s evolution many geographical races havedeveloped. Several localities are regularly fished for thesevariants which have all been given separate names. Afew aquarium favourites received names like “SteveniTiger”, ’Red Empress”, or “Fire Blue”, while others werenamed for the locality where they were collected, e.g.“Steveni Mbenji” or “Steveni Maleri”.

It was known to the fish-collectors that one locationcould harbour two varieties of “Steveni”. “Steveni Ti-ger” and “Steveni Thick Bars” were both caught atLikoma and Chizumulu Island. Although they have dif-ferent preferences concerning their habitat, both Steveniscould be seen together at some locations. “Steveni Ti-ger”, now identified as Protomelas taeniolatus, has thinvertical bars on a silvery body, whereas “Steveni ThickBars” has broad vertical bars. The nuptial coloration ofthe male differs too. A “Steveni Tiger” male has an in-tense blue on the head and upper part of the body and anintense orange-yellow on the flanks. A “Steveni ThickBars” male has a light blue breeding coloration and ayellow throat. Its colours are much less intense com-pared to the “Steveni Tiger”.

From the previous description it must be clear that we

are dealing with two different species which can be foundsympatrically. Ten years ago the scientific identificationof these two species was still unclear. It was Tony Ribbinkand his coworkers (1983) who gave us important newinformation about these two species. They pointed tothe fact that these two species behave differently.Protomelas taeniolatus, the most common of the two,feeds on the biocover in the upper regions of the rockyhabitat. Its food consists mainly of algae. Protomelasfenestratus, the “Steveni Thick Bars”, lives in the deeperregions of the rocky biotope and blows away the sedi-ment on the rocks to reveal its prey, which consists mostlyof insect larvae and crustaceans.

Even though in shape and appearance these two spe-cies are sometimes difficult to distinguish, by observingtheir feeding behaviour one can easily identify them.Ribbink et al. (1983) further noted that the colour pat-tern of P. fenestratus consists of conspicuous verticalbars whereas that of P. taeniolatus has more horizontalelements. Now, with knowledge of behaviour and col-oration it must be easy to identify all populations of thesecichlids in the lake. Initially, it seemed to work. I could(I thought) identify every population even though I foundthat the pattern of markings, especially in P. taeniolatus,varied considerably (which was also mentioned byRibbink et al.).

35

The algal-garden denotes the centre of the territory of thismale P. taeniolatus (Thumbi West Island).

The “Steveni Eastern” is a race of Protomels taeniolatus.

A male Protomelas sp. “Steveni Blue Black” (Masinje). A Protomelas sp. “Steveni Black-Belly” at Mdoka.

After more than 250 hours of diving experience in thelake, the identification of these species looks less sim-ple. P. taeniolatus is very common and territorial malesare found on almost every rocky shore. P. fenestratus ismuch rarer but can be found at most rocky areas as well.Breeding males are not common. They are weakly terri-torial and usually construct a shallow nest in the sandalongside a rock. At Hora Mhango, on the northwesterncoast of the lake, P. fenestratus was found to defend aterritory on top of a rock, like P. taeniolatus does. I alsofound two other species which blow in the substrate aspart of their feeding behavior. One of these, P.pleurotaenia, lives in the shallow sandy and vegetatedareas and does not show geographical variation. Its pat-tern consists of two thin, horizontal lines on a silverybody. The other species is closely related to Placido-chromis johnstoni and therefore easily distinguished fromP. fenestratus. As far as I know, a cichlid with a smallmouth, vertical barring, and blowing in the sediment is asure identification for P. fenestratus.

In May 1989 I found at Kande Island (and atMphandikucha Island) a Steveni-type cichlid which hadthe lower half of the body completely black. I photo-graphed another Steveni-type as well but could not col-

lect it. Since the latter specimen showed distinct verticalbars I provisionally identified it as P. fenestratus althoughI didn’t see it blowing in the sand. I examined one speci-men of the black-bellied Steveni and could not find anydistinction between it and P. taeniolatus. I Thereforethought it to be a geographical race of the latter species.

In November 1990 I revisited the northwestern coastand found black-bellied Stevenis at most locations, al-beit in very low numbers. Furthermore I found a (blow-ing) P. fenestratus at Kande Island which had a heavilybarred colour pattern, unlike the one I had seen before.My conclusion is that there are at least three species inthe P. taeniolatus complex. All of these species can befound sympatrically. The black-bellied Steveni is com-mon only at Kande and Mphandikucha Islands. Its col-oration seems to be quite constant along the northwest-ern shores. Territorial males defend a nest on a rock oron the sand beside a rock.

The rocky areas around the Nsinje river delta on theeast coast of Lake Malawi provided me with anotherproblem. In this area the so-called “Steveni Eastern” iscollected. At first, I had identified this cichlid as P.taeniolatus (Konings, 1989). During a later visit I found,north of the river, another Steveni-type. The males were

36

A male Protomelas fenestratus at its shallow nest in the sand (Nsinje River Outlet).

A female with an unusual colour pattern but identified asP. fenestratus because of her particular feeding technique.

completely blue (see photo) and stayed in the upper re-gions of the habitat. At the same time, at deeper levels, Iobserved P. fenestratus females blowing away in the sand.I then concluded that the “Steveni Eastern” must be P.fenestratus, as I found males at deeper levels than theall-blue Steveni. Moreover I saw a male “Steveni East-ern” displaying for such a sand-blowing female. My onlyreservation at that time was that the females, which werenormally shipped with “Steveni Eastern”, didn’t showmuch vertical barring whereas the sediment-blowingfemales had distinct vertical bars. I never questioned thefishermen about this matter, which I should have done.In a later publication (1990) I showed the “Steveni East-ern” as P. fenestratus.

During my third visit to the area I found territorialmales of P. fenestratus which looked like anything but“Steveni Eastern”. The coloration of these males is verymuch like that of “Steveni Thick Bars” at Likoma andChizumulu Islands. So there are three species at the east-ern shores as well. “Steveni Eastern” is a race of P.taeniolatus in which the females have few markings ona silvery body. The P. fenestratus has never been ex-ported and thus lacks a trade name.

The third species, in which the males are blue, couldin fact be closely related to the black-bellied Steveni fromthe northwestern shores. I have not seen females which

could confirm this idea, but the shape of the short snoutresembles that of the northern species. For the time be-ing, however, we will refer to the blue Steveni asProtomelas sp. “Steveni Blue Black”.

A reef north of Chizumulu Island, named Taiwan Reef,harbours only one member of the P. taeniolatus com-plex. The reef consists of large rocks which meet thesandy bottom at great depths. The depth and strong cur-rent, which is commonly observed around the reef, makethe rocks virtually devoid of sediment. The Steveni atTaiwan has very broad, vertical bars but I have not seenthem blowing into the biocover. If it were P. fenestratus

37

Protomelas sp. “Steveni Taiwan” is easily the most beautiful species of the P. taeniolatus-complex.

this could be explained by there being no sediment toblow in (P. fenestratus hardly ever blows in the gravel inan aquarium). After examining several specimens I mustconclude that the “Steveni Taiwan” is not a race of P.fenestratus. Its anatomy is closer to that of P. taeniolatusbut I now think it is better to regard it as a different spe-cies altogether. I don’t like to split up species-complexesinto many different species, because, although it seemsto solve a taxonomic problem, it is like turning your backon the really interesting questions like variation,speciation, and evolution.

Taiwan Reef may be the only isolated place in thelake that has harboured a viable population of cich-lids, even during the low water stands. We may thusexpect to find old species at this reef. Species thatwere probably never in direct contact with their sib-ling species elsewhere in the lake. Speciation maytherefore have come to a stasis long ago. There arenot many purely rock-dwelling cichlids at the reef;there is only one species of the Pseudotropheus ze-bra-complex, one from the Ps. tropheops-complex,two from the Ps. elongatus-complex; there is only oneMelanochromis and there are no species from the ge-nus Labidochromis. Protomelas sp. “Steveni Taiwan”behaves as P. taeniolatus but males defend their nestsat depths below 20 metres. P. taeniolatus lives mostly

in the upper 10 metres at any rocky coast I have vis-ited. Most of the year there is a heavy current aroundTaiwan Reef, sometimes so strong that it is difficultto swim against, even with fins! The only species thatbraves such a current is Pseudotropheus saulosi. P.sp. “Steveni Taiwan” may have been forced by thecurrent to dwell in deeper regions. P. taeniolatus andthe “Steveni Taiwan” may have the same ancestor butdue to the complete isolation from all other popula-tions it has developed into a different species, adaptedto its specific environment. Its relatively large eyecould therefore be an adaptation to the depth at whichit lives.

References

KONINGS, A. (1989) Malawi cichlids in their naturalhabitat. Verduijn Cichlids, Zevenhuizen, Nether-lands

KONINGS, A. (1990) Book of cichlids and all the otherfishes of Lake Malawi. TFH Publications, Neptune,USA.

RIBBINK, AJ., B.A. MARSH, A.C. MARSH, A.C. RIBBINK,and B.J. SHARP (1983) A preliminary survey of thecichlid fishes of rocky habitats in Lake Malawi. S.Afr. J. Zool 18 (3), pp 149-310.

38

The genus Tramitichromis is characterised by the pe-culiar shape of the lower pharyngeal bone. The teethon this bone are all slender and long. The teeth situ-ated at the front are the longest, which is an unusualfeature. These long anterior teeth are further charac-terised by their long tips which are bent backwards.In most other species the anterior teeth are small andtheir tips point forward. The reason for this particulardevelopment in Tramitichromis is not clear, but mostlikely its origin lies in the feeding behaviour or in thetype of food. Another feature of Tramitichromis is thedownward projecting blade of the lower pharyngealbone. The upper edge of this blade runs horizontallyin most cichlids but in Tramitichromis it projects down-wards (more than 50’ in T. variabilis).

The lower gill-rakers in Tramitichromis are shortand robust, as in other sand-sifting Lethrinops, e.g. L.leptodon, L. macrochir. The first two rakers are usu-ally no more than small knobs on the gill-arch. Thecentral three to five rakers on the lower arches arewide and much larger. The tips of the broad rakerstogether form an almost horizontal platform at thebottom of the buccal cavity. The kind of grid they formmay separate the heavier material from the lighterwhen a mouthful of sand and sediment is taken intothe mouth. The heavier sand may thus sink between

Tramitichromis lituris (Trewavas, 1931)Ad Konings

A territorial male Tramitichromis lituris (Mdoka).

the rakers and be carried to the outside while the lightermaterial, including invertebrates and algae, remainsinside the mouth. The strength of the rakers is prob-ably needed to withstand the abrasive action of thesand. If the rakers were higher, the separating effectmight be more effective. The need to keep the tips ofthe pharyngeal teeth in the same plane as those of thegill-rakers could have led to the situation inTramitichromis where the anterior teeth are relativelylonger than in other cichlids. The anterior pharyngealteeth of e.g. Lethrinops leptodon are not long and arebelow the level of the raker-tips.

I recognize five, possibly six species, in Tramiti-chromis. Four species are described: T. brevis, T.variabilis, T. lituris, and T. trilineatus. Because of itsdifferent colour pattern I regard intermedius – placedin Tramitichromis by Eccles & Trewavas (1989) – asa member of the genus Trematocranus.

I have never been able to examine specimens of T.trilineatus, so all information given here relates to thefirst three species. T. brevis was exported as’Lethrinops Variabilis” or as ’Lethrinops Chizumulu”and is easily recognized by its small adult size (about14 cm) and by the prominent diagonal stripe on itsflank.

The type material of T. variabilis consists, in my

39

A male T. brevis at its cave-crater nest. (Chizumulu).

A Tramitichromis lituris nest on top of a rock.

“Lethrinops Red Flush”, possibly Tramitichromis variabilis, is collected in Senga Bay.

A female Tramitichromis lituris (Mdoka).

A cave-crater nest of Tramitichromis lituris.

40

The lower pharyngeal jaw of Tramitichromis lituris. A T. variabilis-type lower pharyngeal bone; note thesteeply inclined anterior blade. Photos by Gertrud Dudin.

opinion, of at least two different species. Specimensof one of these are exported as “Lethrinops Red Flush”and they were all collected in Senga Bay. This specieshas only vague markings on the body. Trewavas (1931)reports that the northern races (species?) have (mostly)a diagonal stripe. Such specimens have been collectedaround Likoma and Chizumulu Island.

T. lituris has recently been observed in its naturalbiotope but has never, to my knowledge, been ex-ported. A species which I previously identified as T.lituris, is probably an undescribed, but closely relatedspecies (Konings, 1990). They came from Senga Baybut have, since their first importation, never been seenagain.

I have reported on the breeding behaviour of T.brevis (Konings, 1990) and can add only that breed-ing was also observed in November. Since Stuart Grantcollected males in breeding colour in April, it may bepossible that T. brevis breeds throughout the year. Themain feature of the male’s nest is a small stone under(and around) which a semicircular rim of sand (ox-bow) is deposited (see photo), the so-called cave-cra-ter nest. The female lays her eggs as far as possibleunder the small stone.

A breeding colony of T. lituris was observed nearMdoka at the end of November. Most members of theschool were found between a depth of 7 to 15 metres.Territorial males had constructed a cave-crater nest.Similar nests are constructed by T. brevis although theyuse smaller stones. The breeding colony of T. lituriswas mixed with a breeding colony of Nyassachromiseucinostomus the males of which build shallow nestsin the sand when at deeper (than 7 m) levels. It isknown with regard to N. eucinostomus that some malesconstruct a sand-nest on top of rocks whenever thereis a lack of sites on the bottom. The colony at Mdokahad some males defending their nest on top of a rock.

Not only N. eucinostomus but also T. lituris males hadmade, their nests on top of rocks! Such unusual be-haviour in utaka can be explained by the observationthat its females feed on plankton in the open waterand the males are thus closer to the females (even ifthey have to carry the sand up the rock). Lethrinopsforages in the sand and males building a nest on top ofrocks would diminish their chances of mating. AtMdoka, however, I found females several feet abovethe substrate and at first I presumed that these wereattracted by the males that had their territories on therocks. When I examined a few preserved specimens Ifound their long guts (2.5 times standard length) com-pletely filled with phytoplankton. Hence it is possiblethat the territorial males have adapted themselves tothe situation where females feed in the open watercolumn due to a plankton-bloom.

It is still remarkable that a bottom-feeding species,which apparently needs the protection (during spawn-ing) of a cave-crater nest, switches to an open nest-type a few metres above the floor. I have observedfemales accepting the alternative nest and spawningwith the occupant. It is not likely that T. lituris breedsthroughout the year as this species was not seen (atthe same location) in May, 1989.

References

ECCLES, DH. & E. TREWAVAS (1989) Malawian cich-lids. The classification of some Haplochromine gen-era. Lake Fish Movies, Herten, Germany.

KONINGS, A. (1990) Book of cichlids and all the otherfishes of Lake Malawi. TFH Publications, Neptune,New Jersey.

TREWAVAS, E. (1931) A revision of the cichlid fishes ofthe genus Lethrinops, Regan. Ann. Mag. nat. Hist.(10), 7; pp. 133-153.

41

Stigmatochromis sp. “Modestus Makokola”Ad Konings

A wildcaught male Stigmatochromis sp. “Modestus Makokola”.

Stigmatochromis sp. “Modestus Makokola” has beenexported recently, albeit in low numbers. As its nameimplies, it is collected at the Makokola Reef, south ofBoadzulu Island in the southeastern arm of the lake. Ithas also been observed at Boadzulu Island.

S. sp. “Modestus Makokola” differs from S.modestus in several anatomical features. Its averageadult size ranges between 16 and 20 cm. Although S.modestus can grow to a size of about 25 cm, mostadult specimens have a length between 14 and 18 cm.The “Modestus Makokola” has a deeper body and thethree spots are more distinct than in S. modestus. Theusually dark brown body coloration of S. modestusobscures its markings but when stressed, three small,round spots are visible. The first spot (suprapectoral)in S. sp. “Modestus Makokola” is elongated and abouttwice as long as in S. modestus. The body coloration ofthe “Modestus Makokola” is silvery rather than brown.

The teeth of S. sp. “Modestus Makokola” are uni-cuspid and not densely packed, which places it inStigmatochromis rather than Otopharynx. I did notobserve S. modestus at Makokola Reef or at BoadzuluIsland, but they were present at Chinyamwezi andChinyankwazi Islands. Although they were not foundtogether, I believe that the “Modestus Makokola” is adifferent species and not a geographical variant of S.

modestus.There is also a difference in behaviour. S. modestus

lives in the rocky habitat where it usually hides in thedark recesses. The dark brown coloration blends wellwith the environment and provides a splendid camou-flage. It is a typical ambush hunter, its prey beingmainly small mbuna. Only during the breeding perioddo males stake out a territory in the rocky biotope,often under an overhanging rocky ledge. Territorialmales gather in small colonies numbering between 5and 15 individuals.

The few encounters I had with S. sp. “ModestusMakokola” revealed a different behaviour in thispredator. During all encounters (only males were ob-served) the fish was out in the open, not hiding amongthe rocks. At Makokola Reef it seemed to be morecommon near the intermediate habitat at a depth ofabout 35 meters. A few times it was seen cruising overthe sand, probably hunting for small, sand-dwellingcichlids which are abundant in that area. At BoadzuluIsland it was seen once. There it was seen huntingover large rocks, behaving a little like a pursuit hunter.Because of its carnivorous appetite it is desirable toaccompany S. sp. “Modestus Makokola” with largercichlids. Apart from that it is fairly easy to maintain ina community aquarium.

42

Dimidiochromis strigatus (Regan, 1922)Peter Baasch

A territorial male Dimidiochromis strigatus photographed in the author’s aquarium.

Dimidiochromis strigatus has a lake-wide distributionbut it is infrequently collected. Observations in theirnatural habitat are rare and are restricted to subadultindividuals with a length of about 12 cm. Reports aboutits feeding behaviour range from pursuit piscivore, orfeeding on invertebrates, to vegetarian feeding onaquatic plants. If one considers the habitat in whichD. strigatus is usually found, i.e. the shallow sandyareas, most of the alleged diets could indeed make upD. strigatus’ dinner. The oblique mouth indicates thatinsects that have been fallen into the water may alsoform a part of its diet. The size of its mouth and theobservations in the aquarium, however, suggest that itbehaves like a ambush predator preferring live foodabove plants.

D. strigatus is irregularly exported as “Haplo-chromis Sunset”. Most exported specimens have alength of about 12 to 15 cm. The largest specimenever caught measured 24.4 cm total length. D. strigatusis laterally compressed, has a rather deep body, andits lower jaw is longer than the upper. The horizontalmid-lateral stripe on the body is a characteristic of allspecies of Dimidiochromis.

Its morphological features resemble those of D.compressiceps although its body is deeper and lesscompressed than that of D. compressiceps, which gives

it a stouter appearance.Wild caught male D. strigatus start colouring up at

a size between 15 and 20 cm. They develop into beau-tifully coloured cichlids with a characteristic red spotbehind the gill cover and a red anal fin. Males defendterritories when they have assumed the breeding col-oration, and behave rather aggressively towards otherinhabitants of the aquarium. Therefore the tank shouldhave a size of at least 150 cm, but 200 cm would bebetter. In the territory the male constructs a shallowcrater with a diameter of about 50 cm and a depth ofabout 5 to 10 cm.

Spawning takes place in this crater-nest. The numberof eggs laid is remarkable. Once one of my femalesreleased 230 fry after the incubation period! Raisingthe fry poses some problems. Besides nitrate-freewater, the fry also need close attention when feeding.They don’t know when to stop eating. Many casual-ties may arise when too much food is given at onetime. At a size of about 2.5 cm they seem to cope bet-ter with the food.

D. strigatus and D. compressiceps should be keptseparated as hybridization between these two specieshas frequently been reported. Juveniles of the two spe-cies are difficult to tell apart.

43

Peter Baasch

Copadichromis boadzulu (Iles, 1960)

A courting male Copadichromis boadzulu; the female is partly visible in the background.

When one looks through the aquaristic publicationsfor mention of Haplochromis or Cyrtocara boadzuluone gets the impression that this species is one of thebest known of the Malawian cichlids. The photographsaccompanying articles, however, show a different spe-cies to the one on this page. The reason is simple. Thosewho have read the latest revision of Malawi cichlids(Eccles & Trewavas, 1989) and the latest aquaristicliterature (Konings, 1989, 1990) know that the cichlidwhich has been sold for years under names like“Boadzulu”, “Hinderi”, “Red Empress” or (better!) as“Namalenje” belongs to a geographical race ofProtomelas taeniolatus. This has left us with the ques-tion of what does C. boadzulu look like.

The “real” C. boadzulu is, as far as I know, a rarecichlid, small populations of which inhabit the sandyregions in the southeastern arm of the lake. Until nowonly two localities are known, one at Crocodile Rockin 5 metre deep water and one at Makanjila Point wherethe sand is at a depth of about 15 metres. Sightings inthe natural habitat are very rare; breeding individualswere not seen.

C. boadzulu belongs to the utaka group and feedspredominantly on plankton in the open water. In cap-tivity it eats all kinds of aquarium foods. Althoughintraspecific aggression remains within tolerable lim-

its, this active cichlid – maximum size is about 14 cm– needs a rather large aquarium. Initially it may be alittle panicky and thus care must be taken to cover thetank properly.

The basic pigmentation pattern shows some verti-cal barring but this is seen mainly in breeding males.The main characteristic is a horizontal mid-lateral linewhich runs from the third vertical bar to the caudalpeduncle. A second horizontal (dorso-lateral) stripe,which is characteristic of Protomelas taeniolatus, isentirely absent. Fully coloured males have a white-blue blaze on the head which continues as a whitemarginal band in the dorsal fin.

Of great interest was the breeding behaviour dis-played by a male in my tank. He constructed a turretof sand with a height of about 25 cm! When placed inanother aquarium the male again made a sand-cone.

Spawning has not yet been observed. The numberof fry released per spawn, however, is rather smalland never consisted of more than 19 babies. The basicpigmentation pattern and the construction of a spawn-ing-cone may indicate that C. boadzulu belongs toNyassachromis, since most other utaka in Copadi-chromis have a pattern of spots and are not known tobuild spawning-cones on the sand.

44

Buccochromis rhoadesii (Boulenger, 1908)Dr. Andreas Spreinat

A male Buccochromis rhoadesii courting a female (in the background). Photo by A. Spreinat.

Only two of the seven species which are currently as-signed to the genus Buccochromis have experienced awide distribution among aquarists. B. lepturus wasimported several years ago under the tradename“Green Lepturus”. A second species was first importedunder the trade name “Haplochromis Lepturus” andlater as “Yellow Lepturus”. Its scientific name wasuncertain for a long time until the revision of theMalawian haplochromines (non-mbuna) by Eccles andTrewavas (1989) was published. Then it became clearthat “Yellow Lepturus” had been described under thename B. rhoadesii.

B. rhoadesii probably has a lake-wide distributionas it has been caught at several localities around thelake. Most of the exported specimens were collectedat Likoma Island. Here young solitary individuals orsmall groups are regularly seen over mixed sand-rockhabitats but also in rocky areas. They are seen in shal-low water as well as at depths of 15 to 20 metres.

Our B. rhoadesii, which we obtained at a size ofapproximately 6 to 7 cm, were placed, following ashort acclimatization period, in a large tank (250 x 70cm) together with some other haplochromines. Firstwe noticed that these juveniles showed a voraciousappetite. After a year of maintenance the fish hadreached a size of about 16 cm but did not show any

sign of the sexual dimorphism known from adults. Atthis stage they all had a yellow coloration which wasmost intense on the ventral part of the body. At an ageof almost two years and at a length of about 18 cm thelargest specimens began to show signs of the maleadult coloration. The first sign was a blue sheen onthe skin between the eyes and mouth which progressed,with age, to the anterior part of the head and, later on,over the entire body. The yellow coloration decreasedconcomitantly with the appearance of the blue. WhenI could differentiate between the sexes, I took a pairand placed them in a 3-metre aquarium, while I leftthe other pair in the tank they grew up in.

The complete metamorphosis of the males took sev-eral months. The male in the photograph measures 27cm while the females, of the same age, measure be-tween 20 and 23 cm.

Notwithstanding its size and robust appearance, B.rhoadesii behaves relatively peacefully. Although themales need an area of the aquarium in which they areusually found, they do not systematically defend thisas a territory. Females are courted but left unharmed;even during spawning other species are tolerated inthe tank. One thing didn’t, change: their voraciousappetite; thus we expect further growth of our fish.

45

Dr. Andreas Spreinat

Corematodus taeniatus Trewavas, 1935

A territorial male Corematodus taeniatus photographed near Fort Maguire. Photo by A. Spreinat.

The first Corematodus taeniatus were exported byNorman Edwards under the trade name “HaplochromisJacksoni”. In 1988 Stuart Grant exported this cichlid,which was collected at Namalenje Island, as“Haplochromis Space Mouth”. In November that yearI received a call from an importer in Bayern that sev-eral of these cichlids had arrived, which had immedi-ately aroused my interest. One week after the call Ireceived only a single specimen of the ’hew” cichlidwhich had a length of 11 cm. It had a silvery colora-tion and a partially interrupted, diagonal line. It had avery wide mouth in which many rows of teeth werearranged to form a rasp. It was beyond doubt that thiscichlid was a member of the genus Corematodus, tobe precisely C. taeniatus.

The other species which is assigned to this genus isC. shiranus, a cichlid with faint vertical barring. Bothspecies are known to employ a rather unusual forag-ing technique; they scrape small scales from the pe-duncles and caudal fins of other fish, mainly cichlids.

After a short acclimatization period I placed thescale-eater in a community tank with several mbunaand larger haplochromines. Already a few hours afterits introduction the new inhabitant eyed with greatinterest the caudal fins of the other fish. The attacksmostly occurred from behind. With a feigned disin-

terest it would slowly approach its prey followed by asudden bite in the caudal fin. Less common were at-tacks initiated swimming above the victim. The finitself was rarely damaged but the fine scales that covermost of the haplochromine’s fins made up for a seem-ingly nutritious meal. After a few attacks, however,the fin would show rashes and after a few days allother inhabitants of the tank had scrubbed fins. Mean-while, the scale-eater showed a healthy appetite forthe regular aquarium fare as well.

After a while the other fish gradually learned howto escape the attacks and kept a distance to the villain.Especially the mbuna proved to learn quickly. It tookabout two months before an acceptable balance wasreached. The fins of the non-mbuna were healed andthe scale-eater had switched almost entirely to thestandard aquarium food.

When I introduced a few new, unexperienced cich-lids in the tank they were immediately attacked by C.taeniatus. Since they were the only fish in the tankunaware of the scale-eater’s habits they succumbedthe repeated attacks a day later.

In principle it is possible to house C. taeniatus to-gether with other species. Its best tankmates are mbunaor other robust species like those of the genusNimbochromis.

46

Aulonocara rostratum Trewavas, 1935Edwin Reitz

A wild caught male Aulonocara rostratum in breeding coloration. Photos by Edwin Reitz.

Until recently Aulonocara rostratum, which has a lake-wide distribution, was offered as A. macrochir in thetrade, but it is now clear that both these names applyto the same species, i.e. A. macrochir is a synonym ofA. rostratum. Until 1990 only males were exported asfemales were difficult to find. Then, at the beginning of1991, the first few females were exported to Europe.

After two to three weeks of acclimatisation, A.rostratum, like most Malawian cichlids, eats any typeof aquarium fare and starts showing its splendid col-oration. Although it can grow to a size of about 20cm, it has a very small mouth. In the wild A. rostratumfeeds in the typical Aulonocara fashion: it lowers itshead to just above the substrate and tries to locate anyprey moving in the sand. Such behaviour is only ob-served in some specimens in the first few weeks afterimportation.

A. rostratum lives over the sandy substrates of thelake where adult males dig their nests. A breedingcolony of this cichlid contains several males and allof them will have constructed a deep crater-nest whichis defended energetically. Therefore one should giveA. rostratum a large tank with plenty of sand on thebottom. Interestingly, in the aquarium A. rostratumdoes not seem to construct a crater-nest. I have neverseen it digging. What it does is to take away any ir-

regularity from the spawning-site. While doing so itcarries small bits of sand out of its nest each time. Inthe long term the spawning-site becomes a shallow, cleandip in the bottom with a diameter of about 40 cm.

The pre-spawning activity lasts much longer thanin any other known Peacock. With my fish it lasts abouttwo to three days before the actual spawning takesplace. In the meantime the male continuously leadsthe ripe female to its nest.

The male induces the female to deposit some eggsby dragging its lavishly ornamented anal fin over thenest while the pair circle around each other. When thefemale deposits some eggs, up to 10 at a time, themale is parallel to the female (not in T-position). Thefemale turns around quickly and starts picking up theeggs. Meanwhile the male has moved near the eggsand may fertilize them while they are being pickedup. The male makes another pass and again drags itsanal fin over the nest, By now the female has col-lected all of the eggs and picks at the spots on themale’s anal fin. Then the cycle is repeated. A spawn-ing lasts about one hour.

During the incubation the female does not eat. Af-ter three weeks she releases her fry; in small femalesthey number about 50 but in large, well-fed femalesspawns of about 150 fry are common.

47

The female deposits up to 10 eggs at the time. �

After the eggs are laid, the female turns around and picks them up.

The male drags its ornate anal fin over the spawning-siteand entices the female to lay more eggs.

48

Taeniochromis holotaenia (Regan, 1922)Peter Baasch

A male Taeniochromis holotaenia in breeding coloration. Photo by Peter Baasch.

Aquarists surely assess a cichlid in a different way tothat used by the experienced ichthyologist, and try to“arrange” their fish according to personal points ofview. The aquarist has “classified” his cichlids accord-ing their behaviour, gross morphology, and colourpatterns. Other criteria were not available to the lay-man, while scientists applied totally different meth-ods. Both scientist and aquarist had problems appre-ciating each other’s standpoint. The aquarist could notunderstand the reasoning of the scientist, and the sci-entist underestimated the nonscientific approach of theaquarist. The latter could not understand that one hasto measure carefully and count certain features beforea classification of the particular fish could be achieved.

The revision of Malawian haplochromines by DavidEccles and Ethelwynn Trewavas (1989), and the pub-lications of scientists who are aquarists at the sametime, have resulted in characters like morphology, pig-mentation pattern, distribution, and evolutionary traitsbecoming accessible to the hobbyist. The developmentof this mutual understanding may pave the road to abetter cooperation between the scientist and aquarist.

Taeniochromis holotaenia is one of those specieswith a unique pigmentation pattern which is crucialfor its classification. It is the only species in its genus.Although a few specimens had been exported through-

out the years it had never been recognized as an inter-esting cichlid. Now that aquarists are more aware ofthe complexity of the Malawian species flock a re-newed importation of this species in September 1990was greeted with keen interest.

From a morphological viewpoint T. holotaenia be-longs to a group cichlids classified by aquarists as“Torpedos”. Recently such cichlids have been exportedmuch more frequently than before. Usually they areput together in an importer’s tank and labeled“Haplochromis Torpedo”. Among these “Torpedos”there are species from different genera, so careful se-lection from such collections is needed. Cichlids whichare traded as “Torpedos” are e.g. Sciaenochromis gra-cilis, Sc. spilostichus, Stigmatochromis sp. “Spilo-stichus Type”, Champsochromis spilorhynchus, Ch.caeruleus, and Maravichromis formosus. All thesespecies are characterized by a very slender body anda diagonal line, which in some cases consists of a rowof smaller spots, on the body. Among this mixture T.holotaenia is easily identified since it lacks the diago-nal stripe.

T. holotaenia (holo = complete, taenia = stripe) ischaracterized by a horizontal band which connects theeye with the caudal peduncle. In addition to this dis-tinct band a black bar runs horizontally between the

49

T. holotaenia in its natural habitat in Senga Bay (above)and in the aquarium. Note the difference in coloration.

eyes, which gives the impression of the fish beingcompletely circumscribed by a black line. Dependingon the fish’s, emotions vertical bars may be present aswell. This unique and remarkable pattern formed thebasis of placing this species in a genus of its own,Taeniochromis, although it has some resemblance tospecies of the genus Dimidiochromis.

T. holotaenia seems to have a wide distribution inthe lake but it also seems to be rare at most locations.The few observations made in its natural habitat wererestricted to subadults with an average length of about10 cm. They were seen over sand and in the interme-diate habitat at a depth ranging between 10 and 20metres. These individuals all showed a very lightground colour with the distinct black line. Spawningmales, or males in breeding colours, were not seen.The observations, however, indicated that T. holotaeniais a pursuit hunter.

In captivity this cichlid accepts any regular type ofaquarium food. The acclimatization of large adultsdemands a relatively long period, which seems to becommon with piscivores. When kept in sufficiently

large aquariums (over 200 cm) and given regular food,T. holotaenia adapts to its situation and “forgets” aboutits piscivorous nature.

Males become territorial simultaneously with theassumption of breeding coloration. The male con-structs a very shallow dip in the sand as spawning-site, usually near a rock (sometimes even under one).The courting process is rather turbulent. Females witha length of about 20 cm release spawns of about 100fry. At release the fry are small but grow quickly show-ing the characteristic band at an early stage in theirdevelopment.

Juveniles have a silvery-yellow ground colour liketheir counterparts in the lake. Adults, however, showa rather dark pigmentation in the aquarium. Possiblythis colour change also occurs in the lake. I shall haveto wait for the moment the juveniles start changingtheir attire. Male T. holotaenia have a sky-blue breed-ing colour with yellowish fins. In this respect it closelyresembles Sc. gracilis, Sc. spilostichus and St. sp.“Spilostichus Type”.

In the aquarium, compared with other cichlids ofsimilar length and life-style, T. holotaenia behavesrelatively peacefully. Maximum size is around 22 cmfor males.

50

Lethrinops micrentodon (Regan, 1922)Peter Baasch

A wildcaught male Lethrinops micrentodon from Makokola Reef.

Superficially this cichlid seems to belong to the genusAulonocara rather than to Lethrinops. The latter ge-nus includes many species which are aquaristicallyunknown but is getting more and more attention fromaquarists. Eccles and Trewavas (1989) revised thegenus Lethrinops and described two new genera:Taeniolethrinops and Tramitichromis. Although thereis now a certain logic to the systematics of this largegenus, for the aquarist distinguishing the different spe-cies is still problematic. The females in particular pro-vide a major problem because responsible breederswant to prevent hybridization by placing the “right”female with a male.

The species depicted in the photograph above hasbeen exported only once, in spring 1990, and was iden-tified as Lethrinops micrentodon. It belongs to a smallgroup of deep-living species which have a special-ized feeding behavior. Together with L. microdon andL. stridei it feeds predominantly on diatoms whichhave settled on the bottom of the lake. Most speci-mens of these three species have been collected be-low a depth of 50 metres and all three seem to be re-stricted to the southern part of the lake. The L.micrentodon that have been exported were collected atMakokola Reef, south of Boadzulu Island. They werefound at a depth of 36 m at the border of sand and rock.

A few weeks after I had acclimatized my fish tothe confines of the aquarium a male started to stakeout its territory in one of the corners. The territorywas demarcated by a wall of sand which had the shapeof a quarter circle. The radius measured about 45 cm.If enough room had been given the entire territorywould have been a complete circle with a diameter ofalmost one metre! The edge of the construction is justa few centimetres high. In the centre of the territory –in the case of a quarter circle in the corner – the malebuilds a small cone of very fine sand. This cone servesas the spawning-site.

It is not known whether L. micrentodon breedsthroughout the year or if it has a specific breedingseason. The specimens in my tank have never stoppedbreeding since I introduced them into the aquarium,but this behaviour is also known from other Lethrinopsspecies which breed seasonally in the wild.

Females, which reach a maximum size of about 10cm, produce 50 to 60 fry per spawn. The fry are verysmall at the time of release but grow rapidly. At a lengthof about 5 cm the males start colouring up. The maxi-mum size of the males lies around 13 cm. L. micrentodonis a very attractive cichlid. It requires a small aquarium(minimum 100 cm) with a sufficient amount of fine sandto provide material for the construction of a nest.

51

Labeotropheus fuelleborni Ahl, 1927Peter Baasch

A territorial male Labeotropheus fuelleborni at Katale Island near Chilumba.

Several geographical variants of both species ofLabeotropheus are regularly exported and are amongthe popular cichlids from the lake. These two species,L. fuelleborni and L. trewavasae, live sympatricallyat many locations, but hybridisation between themhardly ever occurs. I have seen a hybrid between thesetwo species in the population at Chidunga Rocks (nearChipoka) only. As far as I know there are only tworegions where males of both L. fuelleborni and L.trewavasae are entirely blue. These localities are MaraRocks and the rocky shores near Nkhata Bay. At allother locations only one of the two species is presentor the two species are differently coloured. Most rockyreefs are inhabited by L. trewavasae only whereas atLikoma, Chizumulu, Kande, Mbenji, Namalenje andMumbo Island only L. fuelleborni is present. The rockyshores along the main coasts harbour both species.

There are three major male colour patterns knownfor each species. The most common for L. fuelleborniis entirely navy-blue with distinct barring in territo-rial males. L. trewavasae is also known in completelyblue races but the variant at Nkhata Bay is cobalt-blueand may thus be distinguished from the navy-blue L.fuelleborni at the same location. The second colourpattern present in both species is a blue body and anorange or red dorsal fin. The third pattern consists of

a yellow, orange or rusty brown coloration on theflanks, belly or dorsal part of the body. Some popula-tions of L. fuelleborni have a yellow or orange col-oured belly or flank (see photo above of the KataleIsland population). At one location both species mayhave an orange colour on the body. In L. trewavasae thiscolour is restricted to the upper half of the flank whereasin L. fuelleborni it is present on the lower half.

These three colour patterns are a part of the geneticvariation of the two species. The basic colour seemsto be entirely blue as this is the colour seen in almostall regions where only one of the two species is present.The presence of one of the other two patterns seemsto have originated in a random fashion. It probablyhelped to differentiate between the two species whenthey came into contact (or it became the only way todifferentiate). At each location the process of chance-expression of a pattern and the subsequent sexual se-lection developed into today’s checkered distributionof the various colour patterns. The colour pattern ofL. trewavasae varies more abruptly between adjacentpopulations than that of L. fuelleborni. L. fuelleborniis able to cross sandy areas in shallow water along theshoreline. L. trewavasae is restricted to the deeper partsof the habitat and may thus never have genetic con-tact with nearby populations.

52

Pseudotropheus saulosi Konings, 1990Ad Konings

Although large schools of P. saulosi occur, only a small part of the reef is inhabited by this mbuna.

A territorial male Pseudotropheus saulosi.

The reef north of Chizumulu Island, the so-called Tai-wan Reef, may prove to be the most important placein Lake Malawi to study the evolution of the Malawianspecies flock. While most rock-dwelling cichlids haveformed species complexes at most other locations,most of these complexes are represented by one spe-cies only at Taiwan Reef. The formation of a speciescomplex elsewhere is probably due to the process ofgeographical isolation caused by the fluctuating wa-ter level of the lake. Due to its isolated position suchprocesses probably did not occur at Taiwan Reef. Ifone explains the species complexes by the process of

sympatric speciation (the creation of a new species inthe presence of its ancestor), then how do we explainwhy that process obviously didn’t occur at TaiwanReef. Food is not a limiting factor (at present), be-cause the reef is inhabited by thousands of rock-dwell-ing cichlids and is a rich fishing ground for utaka as well.

Pseudotropheus saulosi is one of the commonestcichlids on the reef and certainly the most conspicu-ous. Females and non-territorial males forage in largeschools of sometimes more than 100 individuals.These troups move in the upper reaches of the reefwhile picking at the biocover. In November 1990 Icould not find P. saulosi below a depth of 14 metres.In May 1989, it was the only species around in theupper regions of the reef while a very strong currentswept over the reef. Territorial males defend a rela-tively large area in front of a cave, but spawning alsooccurred on the open substrate.

P. saulosi shows a strong sexual dichromatism, i.e.males and females have different coloration. The malesstart changing to their blue, zebra-like dress when theyare at a size slightly short of 5 cm (2”). Their maximumsize lies a little over 8.5 cm (3”). In the aquarium, how-ever, they may grow to a size of about 10 cm. Due to itssmall size and vibrant coloration P. saulosi may becomeone of most popular cichlids from Lake Malawi.

53

Labidochromis sp. “Hongi”Peter Knabe

A mature male Labidochromis sp. “Hongi” claims much of the space in the aquarium when kept with other males.

In 1990 during a collection expedition in MbambaBay, Tanzania, Hans Fleischer and Thomas Engel dis-covered a beautifully coloured Labidochromis atHongi Island which they were able to bring back aliveto Germany.

In January 1991 I was able to observe Labido-chromis sp. “Hongi” in its natural habitat. Hongi Is-land, situated between Liuli and Mbamba Bay, con-sists of two smaller islands which contain mainly verylarge boulders. These two islets are separated by a 40m wide and 8 to 12 m deep stretch of water. The biotopeconsists here of a sandy bottom with solitary rocks.Because the two islets consist of large boulders, largecaves and overhanging ledges are frequently foundunderwater. In this type of habitat I was able to ob-serve L. sp. “Hongi” at all levels of the rocky coast.Most individuals, however, were found at a depth ofabout 8 metres. There I saw 8 individuals of whichthree were males displaying in front of a cave. As faras I could observe they fed on plankton or nibbledform the biocover.

I also found L. sp. “Hongi”, albeit at a lower den-sity, at Lundo Island. Lundo is, from an aquaristicviewpoint, a very interesting Island which lies betweenHongi and Mbamba Bay. The shores of Lundo Islandconsist of small – maximum football size – rocks.

In the aquarium L. sp. “Hongi” behaves like mostMbuna. The two males and two females which I ob-tained from the first import were placed in a 1000-litre aquarium. Within one week the more dominantmale spawned with both females. The spawning tookplace on the gravel at a rather randomly chosen site.The number of fry released after three weeks variedbetween seven and 18. The fry grow slowly and reachmaturity after one year. The parents had a length ofabout 6 cm when I obtained them; after almost twoyears of maintenance in captivity the largest male hasgrown to a length of about eight centimetres. The deco-ration of the aquarium should have plenty of shelterfor the females.

L. sp. “Hongi” in its natural habitat. Photo by Peter Knabe.

54

VICTORIANCICHLIDS

Part I: Introduction to taxonomy and ecology

Ole Seehausen

A male “Haplochromis” nyererei, a zooplanktivorous rock dwelling cichlid.

Although the family Cichlidae is otherwise well-docu-mented in aquaristic literature, Lake Victoria cichlidsare one of the few groups which have been given littlecoverage.

For many decades scientific interest in these fisheswas also limited (Barel et al., 1991) and only a hand-ful of ichthyologists were familiar with them. The tax-onomy of Lake Victoria cichlids is still an unresolvedproblem. Regan, in his 1922 revision, covered 48haplochromine species. Greenwood, in his 1979/80revision, knew about 106, and he redistributed theformer Haplochromis members over 15 new and re-

erected genera. Since then about 200 more specieshave been scientifically discovered (most of them stillawaiting description), many of; which bridge the gapsbetween the new genera (Witte, 1974). For these andother reasons (see below) many ichthyologists in theeighties preferred to refer all species back toHaplochromis as an interim measure.

Taxonomic classification in haplochromine cichlidswas traditionally based on skull morphology and den-tition. Striking similarities were found with cichlidspecies from the other African Great Lakes, mainlythose from Lake Malawi, exemplified by pairs of gen-

55

A male “Haplochromis” chilotes, a large species with lobed lips that feeds mainly on insects. Photo by Ole Seehausen.

era like Macropleurodus (Victoria) and Chilotilapia(Malawi) (Regan, 1922) and Paralabidochromis (Vic-toria) and Labidochromis (Malawi) (Greenwood,1956a). Such observations resulted in discussions onpolyphyletic versus the previously assumed mono- oroligo-phyletic nature of the lake flocks (i.e. Green-wood, 1981).

In the last decade molecular biochemical work onEast African cichlids was begun and provided strongsupport for the idea of a monophyletic origin for theVictorian haplochromines (Sage et al., 1984; Meyeret al., 1990), with two exceptions being the non-en-demic species Pseudocrenilabrus multicolor andAstatoreochromis alluaudi. Besides the species ofLake Victoria, the Victorian species flock includes thehaplochromines of Lakes Nabugabo, Kyoga, and veryprobably Lakes Edward, George, and Kivu (cf. Meyeret al., 1990; Greenwood, 1981 for a discussion). Thisflock appears to be the sister group of the Malawianflock and the two together apparently represent thesister group to Astatoreochromis (Meyer et al., 1990).However, the taxonomy at lower levels remains un-clear because of the extremely low genetic variabilitywithin the entire Lake Victoria flock (Meyer, pers.comm.), The striking feature in the evolution of theVictorian haplochromines is that the morphological

diversity is not reflected in a profound genetic diver-gence. Taxonomy is further complicated by a stronggenetically independent variability.

How much do we think we know about the tax-onomy and phylogeny (descent) of Victorianhaplochromines?

1. We have strong evidence to assume a mono-phyletic origin for the species flock.

2. Although species differentiation at this evolution-ary level is not only difficult in practice, but also afundamental problem (Barel et al., 1991), we know ofmany cases in which morphologically and geneticallyalmost identical, sympatric, and parasympatric forms,often differing only in male breeding coloration, areecologically distinct species. In several cases it hasbeen possible to show their ecological segregation(Hoogerhoud et al., 1986; Goldschmidt et al., 1990).However, this is much more difficult in isolated popu-lations.

Why the hesitation to use Greenwood’s new gen-era? The traditional haplochromine taxonomy is to alarge extent based on the anatomy of the feeding ap-paratus. The members of the genus LabrochromisRegan, 1920 (re-erected by Greenwood in 1980), forexample, share a heavy pharyngeal apparatus as theonly derived feature. This was considered indicative

56

of their monophyletic origin (Greenwood, 1980). Eco-logical fieldwork and experiments in the laboratoryhave clearly shown that (1) the elements of the feed-ing apparatus are phenotypically influenced by envi-ronmental factors (eg. type of prey) (Witte, 1984;Hoogerhoud, 1986), (2) many groups described on thebasis of the anatomy of the feeding apparatus and theconnected skull architecture are in effect ecologicalgroups. This does not necessarily deny their validityas taxonomic units but it is quite possible that parallelevolution has occurred, producing superficially simi-lar species whose derived characters are convergentrather than synapomorphic (features with an identicalorigin which species of the group have in common).Another reason for the hesitation is that many specieshave morphological characters which fall in betweenthose of two or more of Greenwood’s lineages mak-ing it impossible to delimit genera (cf. Hoogerhoud,1984; Witte, 1987).

With the background of the extreme difficulties inidentifying and classifying Victorian haplochromineson the one hand, and the increasing need to do so foradequate fisheries management on the other, research-ers from Leiden University worked out the concept oftrophic groups (Witte & Van Oijen, 1990) which al-lows an ecological classification of haplochrominespecies without any phylogenetic implications. Mem-bers of a trophic group are using the same food cat-egory. However, this does not mean that they are feed-ing exclusively on the same prey; food choice can dif-fer seasonally as well as during the 24 hours of theday. To identify the trophic status of a series of speci-mens usually a combination of ecological and mor-phological information has to be considered.

Witte and Van Oijen (1990) further pointed out thatmorphological characters correlate more closely withthe way in which the food is collected than with thefood type itself. Molluscivores, for instance, whichcrush snails between their pharyngeals, are morpho-logically much closer to some insectivores which alsoneed a strong pharyngeal apparatus, than to othermolluscivores which pull snails out of their shells withtheir oral teeth. In such cases Witte and Van Oijendeparted from the strict trophic classification and rec-ognized subtrophic groups on the basis of feeding tech-niques. However, in many cases the actual manner inwhich the fish obtains its food is unknown, and it isobvious that the insectivores, for instance, will haveto be divided into several subtrophic groups onceenough information is available. It is my intention tointroduce the trophic groups in a series of articles andto discuss parts of the Victorian Haplochromis fauna

on this basis.I believe that trophic/ecological identification of

Victorian cichlids provides the aquarist with informa-tion on his/her fish which is more valuable than thatwhich can be gained from unreliable taxonomic(pseudo)identifications which are often no more thanthe result of guessing. At this point I believe it is nec-essary to stress the fact that identification of Lake Vic-toria cichlids by coloration alone is impossible unlessone is very familiar with the very species in questionand even then mistakes are difficult to avoid unlessother characters are checked.

Lake Victoria is just one of at least six East Africancichlid lakes and, being a shallow water body – maxi-mum depth about 90 m – with a huge surface area ofapproximately 69,000 km² it differs considerably inits geography from Lakes Malawi and Tanganyika.

Comparison of the faunas of the different lakes isan intriguing approach that can provide valuable in-formation. However, such investigations have beenfew (Fryer & Iles, 1972; Greenwood, 1981; Witte,1984; Ribbink & Eccles, 1988) because concrete dataallowing direct comparisons are still scarce, and re-cent work on Lake Victoria has shown that statementsbased on meagre data have produced an incorrect pic-ture in the past (Fryer & Iles, 1972 versus Witte, 1984and Dorit, 1986). Only a few words will be added hereon aspects of endemism, speciosity, and trophic com-position of the flocks.

Until quite recently it was believed that Lake Vic-toria cichlids, in contrast to those of Lakes Malawiand Tanganyika, exhibit neither geographic restrictionnor geographic variation (Fryer & Iles, 1972; Green-wood, 1974). The work of the Haplochromis EcologySurvey Team (HEST) in the south of the lake hasshown that this hypothesis does not hold (Witte, 1984).Several examples of intraspecific geographic varia-tion were encountered (Dorit, 1986; Witte & Witte-Maas, 1987) and only a minority of the species has alake-wide distribution (Witte, 1984).

Fryer & Iles (1972) further emphasized that Victo-rian cichlids were ecologically less restricted and spe-cialized when compared to those of the other two lakes.Recent research proves otherwise. Pronounced eco-logical segregation and habitat restriction was foundin many cases involving the following criteria (afterWitte, 1984): bottom type preference; vertical distri-bution along the bottom profile; vertical distributionin the water column; qualitative differences in foodcomposition, including food size partitioning; quanti-tative differences in food composition; differences infood collection strategy; and partitioning of spawning

57

An orange blotched female of a new species from the “H”. nigricans-group, an epilithic scraper. Photos by Ole Seehausen.

“H”. nigricans from an offshore, deep-bodied population,representing the biter type in Haplochromis.

“H”. cf. altigenis, one of the larger piscivores which are almostextinct, representing the sucker type in Haplochromis.

areas. I would like to add behaviour (Seehausen,1991a).

Even within the geographically very restricted areaof a single bay (Van Oijen, 1981; Barel et al., 1991;pers. obs.) one encounters completely different spe-cies assemblages correlated with substrate type andvegetation. I believe that the comparisons of the spe-ciosity of the Great Lakes species flocks which havebeen published over the years merely reflected the stateof knowledge at the time of publication. According toRibbink & Eccles (1988) and Konings (1989) about400 haplochromine cichlids are known from LakeMalawi. Of these about 50% are rock dwelling forms(Mbuna) with an often very limited geographical dis-tribution. For a comparison with the Victorian faunaone has to consider that the existence of a group ofrock-dwelling cichlids similar to Mbuna in Lake Vic-toria has been known only since 1981 (Van Oijen etal.). At that time at least 16 rock frequenting taxa wereknown from the northern part of the Mwanza Gulfalone. Sampling was carried out at rock stations bySevenster, Bouton, Fermon, and by myself during thelast five years and has revealed many more rock fre-quenting cichlids within this small region of the lake.In addition we collected several apparently endemicforms at small offshore islands outside the gulf.

58

View of a habitat; rocks and papyrus are often adjacent.

Ribbink and Eccles (1988) found communities of 9 to14 rock dwelling cichlid species at small isolated rockyoutcrops in Lake Malawi. This number correspondswell with our findings at similar places in Lake Victo-ria. While being aware that it is risky to draw conclu-sions from results of geographically restricted work,it appears likely to me that, conservatively calculated,more than 200 rock dwelling cichlid forms – manyhaving a very restricted distribution – are to be ex-pected in Lake Victoria. Ribbink and Eccles (op. cit.)reported circa 180 haplochromine cichlid speciesfound in a Malawian trawl survey of more than 70stations ranging in depth from 18 to 180 m. Van Oijenet al. (1981) found circa 200 species in a Victoriantrawl survey of an area covering 45 km2

in the Mwanza Gulf which has a maxi-mum depth of only 16 m. I can see lit-tle evidence to assume that the Victo-rian fauna was less speciose than thecurrently known fauna of Lake Ma-lawi. Unfortunately it is no longer pos-sible to prove this because of the im-pact of predation by the introducedLates sp. on the flock.

Due to the patchy information acomparison of the cichlid flocks ontrophic level is also difficult. The pic-ture of the trophic composition of theVictorian flock has changed consider-ably since 1922 and this change showsrather steadily continuing develop-ments: a relative decrease in large spe-cies (i.e. piscivores) and a relative in-crease in small pelagic or semipelagicspecies (i.e. planktivores), deep waterspecies and Aufwuchs feeders. Thisprobably reflects the sampling of thespecies, which is affected by the fish-ing techniques employed by research-ers. Taking the expected increase inrock frequenting cichlids (mainly al-gae scrapers and insectivores) into ac-count, the otherwise pronounced dif-ference between the Victorian andMalawian flocks would be reducedconsiderably. However, the high per-centage of piscivores in the original(pre-Lates) community remains an out-standing feature of the Victorian flock.

Much of the public and scientificdiscussion that has arisen about LakeVictoria during the last years relates to

the enormous impact of the introduction of the nileperch (Lates sp.) on the indigenous fauna (Barel, 1986;Witte et al., 1991). It can be said without doubt thatthe vast majority of the sublittoral and open water spe-cies have declined drastically – about 60% of themhave not been found for several years – and that thelittoral fauna is affected to a lesser degree (Witte etal., 1991; Seehausen, in prep.). It is apparent that thespeed and degree of the decline differed between eco-logical groups. Relevant factors are habitat, adult sizeand abundancy of a species (Witte et al., 1991). Somedetails concerning the different cichlid groups will begiven in following articles.

59

View of a rocky habitat under water: grazing “H”. sp. “Velvet Black”. An adult Lates sp, caught by bottom trawl inthe Mwanza Gulf. Photos by Ole Seehausen

BAREL, C.D.N. (1986) Endemische Cichliden des Viktoriasees vordem Aussterben. DCG-info, 17 (3).

BAREL, C.D.N., W. LIGTVOET, T. GOLDSCHMIDT, F. WITTE & P.C.GOUDSWAARD (1991) The haplochromine cichlids in Lake Vic-toria: an assessment of biological and fisheries interests. In:Keenleyside, M.H.A. (Ed.), Cichlid fishes, behaviour, ecologyand evolution. Chapman & Hall, London.

DORIT, R.L. (1986) Molecular and morphological variation in LakeVictoria haplochromine cichlids (Perciformes: Cichlidae).Ph.D. diss. Harvard, Cambridge, Mass.

FRYER, G. & T.D. ILES (1972) The cichlid fishes of the Great Lakesof Africa. Oliver & Boyd, Edinburgh.

GOLDSCHMIDT, T., F. WITTE & J. DE VISSER (1990) Ecological seg-regation in zooplanktivorous haplochromine species (Pisces:Cichlidae) from Lake Victoria. Oikos (Copenhagen) 58; pp.343-355.

GREENWOOD, P.H. (1965) Environmental effects on the pharyn-geal mill of a cichlid fish, Astatoreochromis alluaudi and theirtaxonomic implications. Proc. Linn. Soc. Lond. 176; 1-10.

GREENWOOD, P.H. (1974) The cichlid fishes of Lake Victoria, EastAfrica: the biology and evolution of a species flock. Bull. Br.Mus. Nat. Hist. (Zool.) Suppl. 6.

GREENWOOD, P.H. (1981) The haplochromine fishes of the EastAfrican lakes. Kraus-Thomson Organization, Munich. (Includesall the cited papers in reprint unless given separately)

HOOGERHOUD, R J.C. (1984) A taxonomic reconsideration of thehaplochromine genera Gaurochromis Greenwood, 1980 andLabrochromis Regan, 1920 (Pisces; Cichlidae). Neth. J. Zool.34; pp 539-565.

HOOGERHOUD, R.J.C. (1986) Ecological morphology of some cichlidfishes. Ph.D. Thesis, Leiden, Netherlands.

HOOGERHOUD, R.J.C., F.WITTE & C.D.N. BAREL (1983) The eco-logical differentiation of the closely resembling Haplochromisspecies from Lake Victoria. Neth. J. Zool. 33; pp 283-305.

MEYER, A., T.D. KOCHER, P. BASASIBWAKI & A.C. WILSON (1990)Monophyletic origin of Lake Victoria cichlid fishes suggestedby mitochondrial DNA sequences. Nature 347; pp 550-553.

OIJEN, M.J.P. VAN, F. WITTE & E.L.M. WITTE-MAAS (1981) An in-troduction to ecological and taxonomic investigations on the

haplochromine cichlids from the Mwanza Gulf of Lake Victo-ria. Neth. J. Zool. 31; pp 149-174.

REGAN, C.T. (1922) The cichlid fishes of Lake Victoria. Proc. Zool.Soc. Lond., pp 157-191.

RIBBINK, AJ. & D.H. ECCLES (1988) Fish communities in the EastAfrican Great Lakes. In; C. Léveque, MN. Bruton & G.W.Sentongo (Eds.) Biology and ecology of African freshwaterfishes. ORSTOM, Paris.

SAGE, R.D., P.V. LOISELLE, P. BASASIBWAKI & A.C. WILSON (1984).Molecular versus morphological change among cichlid fishesof Lake Victoria. In: A.A. Echelle & I. Komfield (Eds.) Evolu-tion of species flocks. University of Maine at Orono Press.

SEEHAUSEN, O. (1991a) A comparison of some ethological aspectsof reproductive ecology in three zooplanktivorous and one gen-eralized insectivorous Haplochromis from Lake Victoria. 7thIntl. Ichthyol. Congress (The Hague); Bull. zool. Mus. Univ.Amsterdam, special issue, August 1991.

SEEHAUSEN, O. (1991b) Die zooplanktivoren Cichliden desViktoriasees. DATZ, 44; pp 715-721.

WITTE, F. (1984a) Consistency and functional significance of mor-phological differences between wild-caught and domesticHaplochromis squamipinnis (Pisces: Cichlidae). Neth. J. Zool.34; pp 596-612.

WITTE, F. (1984b) Ecological differentiation in Lake Victoriahaplochromines: comparison of cichlid species flocks in Afri-can lakes. In; A.A. Echelle & I. Komfield (Eds.) Evolution ofspecies flocks. University of Maine at Orono Press.

WITTE, F., T. GOLDSCHMIDT, J. WANINK, M. VAN OIJEN, K.GOUDSWAARD & W. LIGTVOET (1991) Species extinction and eco-logical changes in Lake Victoria. Proc. 7th Intl. Ichthyol. Con-gress (The Hague), in press.

WITTE, F. & M J.P. VAN OIJEN (1990) Taxonomy, ecology and fish-ery of: Lake Victoria haplochromine trophic groups. Zool Verh.(Natl. Nat. Hist. Mus. Leiden) 262.

WITTE, F. & E.L.M. WITTE-MAAS (1987) Implications for taxonomyand functional morphology of intraspecific variation inhaplochromine cichlids of Lake Victoria. In: F. Witte. Fromform to fishery Ph.D. Thesis, Leiden, Netherlands.

References

60

WEST AFRICANCICHLIDS

Lamprologus sp. “Kinganga”

Frank Warzel

The cranial gibbosity of the male Lamprologus sp. “Kinganga” is an interesting feature. Photos by Frank Warzel.

This Lamprologus, for once not one of the many newspecies from Lake Tanganyika, was collected by HeikoBleher during one of his many collecting expeditionsin Africa. The male shown in the photographs, whichis the only specimen of this species collected thus far,was caught in the Lower Zaïre (Congo River) nearKinganga about 200 km downstream from the Zaïreancapital, Kinshasa.

The Lower Zaïre at Kinganga had been the site ofan earlier expedition by ichthyologists about twentyyears ago. An American expedition led by TysonRoberts and Donald Stewart collected many species

of fish here in 1973. Three years later both scientistspublished the results of their expedition and in one ofthe many plates accompanying the publication a smallcichlid is depicted with the caption Lamprologuswerneri. The type species of L. werneri (Poll, 1959:108-109, Pl. XIX), however, was collected in the lowerpart of Malebo Pool, a very shallow, lake-like widen-ing of the lower Zaïre river. Malebo Pool is a frequentlyvisited site where exporters collect their aquariumfishes. As well as Steatocranus casuarius (BuffaloHead Cichlid), which is regularly exported in largequantities from this area, L. werneri, usually under

61

Lamprologus sp. “Kinganga”, hopefully not the only importation.

trade names like “Lamprologus Congoensis” or“Lamprologus Congolensis”, has also found its wayinto many aquaria. Although this cichlid closely re-sembles the “new” Lamprologus sp. “Kinganga”, es-pecially in its cylindrical shape, there are a number ofminor differences which suggest that we are dealingwith a distinct species and not with a geographicalvariant of one species, namely Lamprologus werneri.

The male Lamprologus sp. “Kinganga” that I keptfor several years in one of my aquaria, grew to a lengthof about 10 cm, which is several centimetres shorterthan the maximum length known for L. werneri. Thepattern of light spots on the dorsal and caudal fins ismuch brighter in Lamprologus sp. “Kinganga” thanin L. werneri. Also the vertical markings on the dorsalpart of the body are narrower compared to those in L.werneri. The relatively large cranial gibbosity of the“new” species and the distinct markings on the scalesmay be further, possibly typical, characteristics of thisspecies. Roberts and Stewart noted in their earlierquoted publication that they too found differencesbetween these cichlids, and consequently treated bothforms as possible geographical variants of L. werneri.

As an aquarium inhabitant Lamprologus sp.’Kinganga” is an undemanding, sometimes territorial

cichlid which accepts regular aquarium fare and tol-erates any type of water with no noticeable distress.Even larger fishes find it difficult to intimidate thissmall cichlid; typical Lamprologus.

References

POLL, M (1959) Recherches sur la faune ichthyo-logique de la region du Stanley Pool. Annls. Mus. r,Congo Belge. Ser 8º; Sci. Zool. 71; pp 75-174, pls.XII-XXVI.

ROBERTS, T.R. & D.J. STEWART (1976) An ecologicaland systematic survey of fishes in the rapids of thelower Zaïre or Congo River. Bull. Mus. Comp. Zool.147, N 6; pp 239-317.

WARZEL, F. (1990) Ein neuer Lamprologus aus demunteren Zaire? DATZ (43), 2; pp 74-75.

62

Tilapia tholloni (Sauvage, 1884)Jan ‘t Hooft

A female T. tholloni fanning her eggs. Photos Jan ’t Hooft.

The distribution patterns of T. tholloni (yellow), T. rendalli(red) and T. zillii (blue).

The fact that the African cichlids of the genus Tilapia,and the closely related genera Sarotherodon andOreochromis, are relatively poorly represented inaquaristic literature can be explained by their gener-ally large adult size and often destructive behaviourin the aquarium. Tilapiines, however, are very inter-esting and I will describe my experiences with one ofthe most beautiful among them.

Tilapia tholloni was described by Sauvage in 1884under the name Chromis tholloni, honouring the col-lector Tholon. Boulenger (1899) placed it in the ge-nus Tilapia where it remains today. T. tholloni is wide-spread in West Africa and is found in Gabon, CongoRepublic, the southern part of Cameroun, and in a partof Zaïre. It is also found in the brackish waters ofcoastal lagoons although the water of the rivers andstreams in which most individuals are found containsvery few minerals. Obviously the water chemistryplays an unimportant role.

T. tholloni belongs to the vegetarians in the familyCichlidae. Examinations of stomach contents revealpredominantly remains of plants supplemented withinsect-larvae and crustaceans. Two other species inthe genus have a similar vegetarian diet. These are T.zillii and T. rendalli. These three cichlids have a non-overlapping distribution in most parts of Africa (Thysvan den Audenaerde, 1963; see map).

The specimens I kept in the aquarium were collectedby friends of mine in the Congo Republic. The juve-niles, with a length of about three centimetres, werecaptured in a swampy area just south of the AlimaRiver, a tributary of the Zaïre (Congo), not far fromthe village of Oyo. The water had a depth of 30-60 cmand the small tilapias could be collected with a handnetfrom among the grassy vegetation. After I had intro-duced four of them in a metre-long aquarium theyimmediately showed their vegetarian nature. Within afew days they had completely eaten a thick layer ofduckweed which had covered the water in the tank.Then they started, selectively, on the other plants inthe tank. Some plants like Anubias, Ceratophyllum,and Nuphar lutea were left untouched, but soft-leavedplants were devoured in days. I fed them lettuce leavesand from time to time mosquito larvae and Mysis.

The juvenile T. tholloni had a silvery coloration withan occasional horizontal line of small spots. At a fewmonths old, at a size of about six to seven centimetre,they changed colour. A brown colour with a greenishhue covered the once silvery body. They were thenplaced in a 200 cm-long aquarium in which theychanged colour again at a size of about 10 cm and anestimated age of about nine months. The change con-

sisted of the appearance of a red colour on throat andbelly. Shortly after this colour change one of the fourshowed a further, this time drastic change in colourand behaved territorially. Its territory included morethan half of the tank. A few days later a larger indi-vidual was allowed in its territory and this specimenalso adopted the breeding coloration.

Previous reports (Schuitema, 1963; Nieuwenhuizen,1967) describe the construction of large spawning pitsin which eggs are laid, but in my tank the pair spawnedbefore a pit was dug. One day after the second indi-vidual entered the territory the female had attached alarge batch of eggs onto a piece of bogwood. The pairtook turns fanning the eggs and dug a large pit. Fourdays after deposition the eggs hatched and the larvaewere removed to the pit. The wriggling mass was pro-tected by both parents.

In the meantime the remaining two fishes formed apair as well and spawned within a few days of the firstpair. This pair too started digging after the eggs weredeposited. The spawns of these young couples num-bered more than 500 fry. For months on end the two

63

A pair of wild caught Tilapia tholloni guarding their fry.

An adult T. tholloni with neutral coloration.

A juvenile T. tholloni showing its silvery coloration.

pairs spawned. Eggs were always stuck onto a smoothvertical substrate and the nursery-pit was always dugone day after spawning.

The four T. tholloni reached a size of about 18 cmbefore I gave them away. The maximum size of thisspecies may be close to 25 cm. The question is whetherthis cichlid, under natural circumstances, deposits eggsonto a hard, smooth surface or digs a pit in which theeggs are deposited on the bottom. Maybe the spawn-ing method is dependent on environmental factors andboth methods are practised.

References

NIEUWENHUIZEN, A. van den (1967) Tilapia tholloni.Het Aquarium. 38; pp 158-164.

SAUVAGE, H.E. (1884) Note sur les poissons deFranceville, Haut Ogoaué. Bull. Soc. zool. Fr. 9; pp193-198.

SCHUITEMA, A.K. (1963) Tilapia tholloni. HetAquarium. 34; pp 83-85.

THYS VAN DEN AUDENAERDE, D.F.E. (1963) La distribu-tion geographique des Tilapia au Congo. Bull. Acad.r. Sci. Outre-mer. 9; pp 570-605.

64

Ctenochromis polli and Thoracochromis demeusiiMartin Geerts

A wild caught male Thoracochromis demeusii from Kinganga, Zaïre.

The African cichlid fauna can be split up into two maingeographical groups: a West African group which,aquaristically seen, is dominated by Chromidotilapia-like cichlids and a East African group dominated byhaplochromine and lamprologine cichlids. Greenwood(1987, in Bull. Brit. Mus. nat. Hist. (Zool) 53 (3): 200)wrote that the chromidotilapiine cichlids are adaptedto habitats with flowing water, whereas the haplo-chromines are noted for their trophic specialisations.Although chromidotilapiine cichlids are absent fromEast Africa, the geographical separation of the twogroups is not absolute. This is clearly shown by thecichlid fauna of the lower Zaïre. Here severalhaplochromine cichlids have adopted a rheophilic lifestyle. The species to which these cichlids belong wererevised in 1964 by D. Thys van den Audenaerde, whorecognized four different species: Haplochromisfasciatus, H. demeusii, H. bakongo and H. polli. Inhis subsequently contested revision, Greenwood(1979) placed the first three species in Thoracochromisand the last in Ctenochromis. Two of these specieshave been introduced into the hobby.

Ctenochromis polli (Thys v. d. Audenaerde, 1964)was first introduced in the early seventies. At that timeit was known under the name “Pelvicachromis Polli”which suggests that aquarists at that time didn’t ex-

pect haplochromine cichlids in West Africa. Ct. polliis only found in the old basin of Malebo Pool (StanleyPool). Earlier authors have regarded their specimens,which were caught in the Pool, as representatives ofTh. fasciatus. The maximum length of Ct. polli malesamounts to approximately 10 cm. Females remain alittle smaller. The anal fin of the male shows only asingle egg-spot, which was a reason for certainaquarists to experiment with this species in order tofind out the function of the spot. The single egg-spotcould easily be removed, but investigations indicatedthat it played no major role at the fertilisation of theeggs.

Thoracochromis demeusii (Boulenger, 1899), theWest African Humphead, was introduced only recentlyinto the aquaristic hobby. According to Mayland (1989,Das Aquarium 244: 613) the Humphead was collectedby Heiko Bleher in the lower Zaïre near a village calledKiganga. Male Th. demeusii grow to a size of at least14 cm (Mayland, Cichliden Afrikas: 17-18). Adultmales have a cranial gibbosity which led the Britishichthyologist Charles Tate Regan to place this cichlidin the genus Cyphotilapia. Since its introduction it hasbeen spawned in the aquarium. Its breeding behav-iour is typical of a haplochromine mouthbrooder.

65

CENTRAL AMERICANCICHLIDS

The “Cichlasoma” labridens-complex

Juan Miguel Artigas Azas

The waterfalls at Tamasopo form a boundary in the distribution of some cichlids. All photos by Juan Miguel Artigas Azas.

The Pánuco river system, México’s second largestAtlantic drainage and one of the most beautiful foundin the country, has been pouring its entire flow forthousands of years into the Gulf of México. A beauti-ful assemblage of cichlids of the Parapetenia sectionof the genus Cichlasoma are endemic to the Pánucobasin. One of the species is of uncertain descent andhas a restricted range: “Cichlasoma” bartoni from theRio Verde valley in the western part of the basin. Theother species can be placed into two groups, the firstassociated with “C.” carpinte and the second with “C.“labridens.

At present three species in the labridens-complexare scientifically described. “C.” labridens was firstdescribed as Heros labridens by the ichthyologistPellegrin in 1903 from specimens collected by pro-fessor Alfredo Duges of the Mexican University ofGuanajuato in the Rio Verde near the city of the samename. The type locality, however, was stated asHuasteca Potosina, the name of a large part of the RioPánuco basin. Pellegrin also added Guanajuato toHuasteca Potosina as a type locality, something whichis clearly a mistake (there are no cichlid fishes nativeto Guanajuato state).

66

The distribution of “C.” labridens (2, 3, 4), “C.” steindachneri(5), “C.” sp. “Labridens Tamasopo” (5, 6), “C.” sp. “LabridensBlue” (7, 8, 9, 10) and “C.” pantostictum (11).

A natural hybrid between “C.” carpinte and “C.” labridens.“C.” labridens from the Rio Verde.

1. Pánuco2. Rio Verde3. Rio Santa Maria4. Media Luna5. Tamasopo6. Rio Gallinas7. Tampaón8. Tempoal9. Moctezuma10. Tamesi11. Laguna Puerta

In 1983 Jeffrey N. Taylor and Robert Rush Millerdescribed “C.” pantostictum and gave redescriptionsof ““C.” steindachneri and “C.” labridens. Theredescription of the last species was based on a groupof labridens forms which were collected at over fiftydifferent locations in the Pánuco system. Dr. RobertRush Miller (pers. comm.) plans to describe anotherspecies in this group, the one known as the TamasopoLabridens. This labridens lives in the isolated RioGallinas in the mid-western part of the system. Anestimated twelve thousand years ago the valley of RioVerde – about 10,000 km2 of dry land full of interest-ing endemic flora and fauna and situated 1000 metresabove sea level – was the site of a large lake. In thecourse of time, either erosion or tectonic activitycaused this lake to drain to the Gulf, leaving behind alarge swampy area where primitive man is known tohave hunted mammoths.

Over the years the swamp almost dried out, leavingsmall swampy regions scattered over the area. Thesesmall swamps survived thanks to the warm watersprings that feed them. The swamps are isolated fromone other, but contain what used to be the fauna of theprehistoric lake. Two cichlids, namely “C.” bartoniand “C.” labridens, share their natural habitat in thosesprings with several other species of fishes. The latteris the so-called Yellow Labridens.

“C.” labridens feeds exclusively on crustaceans andsnails which it finds in the sand and detritus on thebottom of the springs. It has well-developed, molari-form pharyngeal teeth with which it can crush its preywithout problem.

This cichlid, which has a marked ability to changecolours according to its mood, makes an almost reli-gious ceremony of feeding time by changing its nor-mally yellowish coloration to a dark, sometimes vel-vety black pattern with a sprinkling of blue. But thishabit of changing colours reaches its maximum when

the fish starts breeding, which, due to the very stableenvironment of the springs, takes place throughout theyear. A peak in breeding activity is noted from De-cember to March. A canary yellow and velvety blackcolour pattern then adorns “C.” labridens and thischange makes it one of the most delightful sightsamong the New World cichlids (see photo in CichlidsYearbook, vol. 1: 75).

Pairs then look for a solid surface on which to laytheir eggs. Something that would be an easy task in ariver, but not in the springs where stones are scarce.Frequently stems or leaves of water lillies serve as aspawning substrate. Once a site has been chosen andthoroughly cleaned, hundred of large – about threemillimetres long – orange eggs are laid, fertilized, andfanned for two days. After the eggs have hatched thelarvae are transported to a large pre-dug pit below thespawning site. From then on it takes another five daysbefore the larvae become free-swimming fry. Pairs willthen protect their fry in a ferocious way. The food, inthe form of organic detritus, is provided by the femaleby wagging her entire body in the sediment. Thiscauses a cloud of debris in which the fry will greedily

67

A pair “C.” sp. “Labridens Tamasopo” guarding their fry under the waterfall at Tamasopo.

forage. The pair take the fry on foraging trips throughthe habitat, interrupted every now and then by out-breaks of potential danger in the form of predators ora curious human cichlid lover. At such moments themale will either face the intruder or, if it is too big,flee away. The female will then call her fry and allwill hide under the leaves of water lillies until the dan-ger disappears. At dusk as well shelter is found amongthe lillies. Not until the fry have reached two to threecentimetres in length, do they start making solitaryforaging trips, and gradually those that survive willleave their parents. At this point they stay close to theoverhanging vegetation along the shores of the chan-nels which supplies them with shelter. Juveniles arealso commonly seen around feeding adults where theymay learn the technique of stirring up the sedimentand, while watching, benefit from uncovered smallsnails or other invertebrates that the adults won’t con-sider. “C.” labridens and “C.” bartoni share the habi-tat, ignoring each other most of the time. This is per-haps due to the fact that no direct feeding competitionexists between them. “C.” bartoni, which has conicalpharyngeal teeth, feeds upon algae on the surface ofwaterlilly leaves or other smooth surfaces, or onzooplankton. The ratio between both species, how-ever, is overwhelmingly on the side of “C.” bartoniand, in fact, the yellow labridens from the springs,

although present in all of them, may be consideredrare.

Medialuna spring, the largest of the Rio Verde val-ley swamps, presents additional problems for thefishes. Man-made irrigation canals, called “asequias”,have been dug from this spring since the seventeenthcentury. A large concrete canal that feeds a network ofsmaller ones was built by the government in 1977 totake advantage of the large flow of the spring. Thenew channel intersects the Rio Verde where a differ-ent fish fauna thrives. During the rainy season the rivermay overflow and mix with the spring water in thecanal. In this way fishes from the spring can get intothe river and vice versa. “C.” carpinte has thus colo-nized the canal and, although it does not do very wellin the spring, it has hybridized with the yellowlabridens. Moreover it presents an additional threat tospring cichlids as juvenile carpinte prey upon fry ofthe endemic species.

Sarotherodon species have also been introduced tothe springs as a food fish. It was feared that they couldpose a threat to the naturally occurring cichlids, butwith the passage of time I have observed their num-bers decreasing. In some springs they have completelyvanished.

The Rio Verde labridens is very similar to the onefound in the springs. It differs mainly in body mor-

68

“C.” sp. “Labridens Tamasopo” stirring the sediment. A male “C.” sp. “Labridens Blue” from the Rio Pujal.

A natural “C.” steindachneri x Tamasopo Labridens hybrid.“C.” steindachneri is the largest cichlid in the Pánuco.

phology, having a more rounded profile and a shal-lower head. It has, however, the same breeding col-oration. The habitat in the Rio Verde is rather differ-ent from that found in the springs. While in the springsthe labridens are used to very slow flowing water withstable temperatures from around 28° C to 32° C andgood visibility, the river has moderately fast-flowingwater – the labridens’ preferred habitat – and signifi-cant fluctuations in water level and temperature (from18° C to 28° C). The visibility in the river is rarelyover two metres and food is plentiful.

The “C.” labridens in the river feed on invertebrateswhich they collect in a similar way to their counter-parts in the springs. The breeding season is limited tothe months of March through June. Breeding stopswhen heavy rains cause much higher water levels withan accompanying decrease in visibility and a lowertemperature.

As a spawning site, males of the yellow labridensdig a cave with a depth and width of about 10 to 15cm at the base of a rock, One of the cave’s walls iscleaned and prepared as a spawning substratum. Therest of the breeding behaviour of the river labridensmatches that of their counterparts in the springs.

The yellow labridens is distributed over the west-ern part of the Rio Verde drainage, in the Verde andSanta María rivers. This area is situated partly at analtitude of 1000 m above sea level, upstream of theconfluence of the two rivers. It is also found in thedrainage of this area, the Rio Tampaón, until geo-graphical barriers in the form of waterfalls preventmigration further downstream in the system.

Downstream in the Tampaón river the 102 m highTamul waterfall provides an effective way of isolat-ing a unique fish fauna in the Rio Gallinas. Two cich-lids from the labridens-group live in this stream in apeculiar and close relationship. One of them is “C.”steindachneri, described by Jordan and Snyder in 1900and with a stated type locality of Rio Gallinas nearthe town of Rascon; the other is the so called Tamasopolabridens, a so far undescribed species.

“C.” steindachneri is a large, slender cichlid with along snout and large conical teeth. It is a piscivoreand frequently hunts, livebearers and tetras. On theother hand, “C.” sp. “Labridens Tamasopo” is some-what smaller, with molariform pharyngeal teeth and arounded profile. It feeds on invertebrates and collectsthem in the same fashion as described for “C.”

69

A female “Cichlasoma” pantostictum in breeding coloration (from Laguna de la Puerta).

A male “C.” pantostictum in normal dress. Laguna de la Puerta, the type locality of “C.” pantostictum.

labridens. The estimated ratio between the two spe-cies is about 100 Labridens for each “C.” stein-dachneri. The Rio Gallinas is characterized by its clearand well-oxygenated water (pH 7.6; 100° DH!) andhas a temperature ranging between 16° and 28° C.

“C.” steindachneri can be found only in the RioAgua Buena, between the waterfalls at Tamasopo andthe town of Tambaca, and in Rio Ojo Frio as far as itsconfluence with the Rio Agua Buena, where both riv-ers merge to form the Rio Gallinas. The TamasopoLabridens exceeds that range to the headwaters of theRio Tamasopo, above the waterfalls. This may sug-gest that the labridens arrived before “C.” stein-

dachneri in the Rio Gallinas. Both species breed fromDecember to June with a peak in activity in March.During this period the pairs dig caves between therocks and breed like the yellow “C.” labridens in theRio Verde. The only difference is the colour of theeggs which is yellowish instead of orange. Mated pairsof the Tamasopo Labridens have a minimum size ofabout ten centimetres for males and seven for females.In contrast, “C.” steindachneri forms pairs only whenthe male is larger than twenty and the female largerthan fifteen centimetres.

The breeding coloration of the two species also dif-fers. The Tamasopo Labridens has a black and white

70

chessboard pattern with a white forehead. “C.”steindachneri hardly changes its coloration.

Interestingly, hybrids of the two cichlids exist. Apair consisting of a male steindachneri and a femaleTamasopo Labridens was photographed underwaterby Ad Konings in March 1991. This explains the oc-currence of numerous fishes that have a shape in be-tween the two, and these are only found below theTamasopo waterfalls where both species live together.

“C.” steindachneri is the largest cichlid in thePánuco system, in some cases growing to over 40 cmin size. In some exceptional cases Tamasopo Labridensmales can grow to over 30 cm in length but most largemales average around 20 cm.

Downstream in the Tampaón river and in the low-land drainage of the Pánuco (including the Tamesiriver, the Pánuco’s northern branch, and the Mocte-zuma and Tempoal rivers, the southern ones), lives adifferent form of the labridens-complex which isknown as the Blue Labridens. It is distributed over awide area including some small rivers to the souththat do not belong to the Pánuco system.

The behaviour of the Blue Labridens is almost iden-tical to that of the previous two labridens. It prefersfast-flowing water with rock-strewn floors for feed-ing. The breeding coloration consists of a blue patternand a red edge on the dorsal and anal fins. The scale-less blotch behind the pectoral fin, a common featureof all species in this group, is blood-red instead of thedark-purple observed in all others.

“C.” pantostictum lives in isolated coastal lagoonsnear the Pánuco mouth. Its habitat is very differentfrom those in which the other species of the labridens-group are found. The type specimens were collectedin the coastal lagoons “Laguna de la Puerta” and“Laguna del Chairel” (type locality), both located nearthe city of Tampico. The lagoons are murky with avery low visibility of usually less than half a metre;the bottoms are always muddy. In lagoons with clearwater and sandy bottoms “C.” pantostictum could notbe found. The temperature of the water in the lagoons,which is slightly salty, varies from around 15° C to28° C.

“C.” pantostictum – its specific name means “spot-ted all over” – differs from the other labridens formsby having its entire body covered with small brownspots. Such spots can also be found in the BlueLabridens, but to a much lesser degree. The shape of“C.” pantostictum resembles most closely that of the“C.” labridens of the Rio Verde valley. This may notsurprise us given the similarity of the bottoms fromwhich they feed and the almost stagnant water of their

habitats. The breeding coloration shows the pattern ofthe labridens-complex, but this time is all black witha white forehead. Its breeding habits are difficult toassess because of the murky water in which it lives. Ihave collected “C.” pantostictum in breeding coloursfrom April to June. This cichlid, which grows to over25 cm in length, has already been spawned in captiv-ity by Don Danko (Cleveland, Ohio).

An interesting note about “C.” pantostictum is thattwo specimens were collected in the head waters ofthe Rio Sabinas in the upper part of the Tamesi drain-age. They seemed to live together with the BlueLabridens (Darnell, 1962). Whether they were “C.”pantostictum or somewhat more spotted specimens ofthe Blue Labridens remains to be seen. I could notfind “C.” pantostictum in the Rio Sabinas or in otheraffluents of the Rio Tamesi.

Of all the species in the labridens-complex, in myown experience “C.” pantostictum is the most aggres-sive one. All these species require a tank over fivehundred litres and can be housed with other large“Cichlasoma”. Fishes that are too big to fit theirmouths, except in the case of “C.” steindachneri, arenormally ignored outside breeding times. The largepredator of the group will be pleased with all the smallfish you can provide. “C.” steindachneri is also theleast aggressive of the group.

The foregoing should have given a picture of thecurrent status of the “C.” labridens complex. Manyquestions remain as to whether these forms are to beregarded as different species or as geographical vari-ants of just one. Nevertheless, they provide the aquaristwith interesting fishes with great personality.

References

DARNELL, R. M. (1962) Fishes of the Rio Tamesí andrelated coastal lagoons in east-central Mexico, Publ.Inst. Mar. Sci. 8; pp 299-365.

MEZO, L.M., C. VELASQUEZ, C. ROJAS & J. RIVERA

(1989) Rioverde. Ayuntamiento de la ciudad deRioverde.

PELLEGRIN, J. (1903) Description de cichlides nouveauxde la collection du Muséum. Bull. Mus. Nat. Paris.9; pp 120-125.

TAYLOR, J J. N. & R.R. MILLER (1983) Cichlid Fishes(Genus Cichlasoma) of the Rio Pánuco basin, East-ern Mexico, with description of a new species. Occ.paper Mus. Nat. His. Univ. of Kansas. 104; pp 1-24.

71

One of the many springs near Cuatro Cienegas. All photos of “Cichlasoma” minckleyi in this article were taken here.

“Cichlasoma” minckleyi Kornfield & Taylor, 1983Ad Konings

”Cichlasoma” minckleyi is endemic to the springs atCuatro Cienegas in the northern part of México, inthe state of Coahuila. Cuatro Cienegas lies in the east-ern part of the Chihuahua desert. Due to its geographi-cal location it was isolated from other water systemsuntil men built canals to irrigate land and thus con-nected the springs with the Rio Salado (Rio Grandesystem).

The water system at Cuatro Cienegas consists ofmany springs (thermal and cold) that drain into lakesor form small creeks or rivers. The water is very hard,more than 50° DH, and the temperature in a spring Ivisited in May 1991 was more than 35° C. Originallythere were seven independent drainage systems ofwhich most drained into closed lakes (Minckley, 1969).In 1974 only two drainage systems were not intercon-nected by canals, all others draining into neighbour-ing systems or into the Rio Sa1ado (La Bounty, 1974).Moreover an extensive underground water systemconnects several springs with each other and throughthese underground channels fish may have migratedfrom one system to another (Minckley, 1969). Thebasin of Cuatro Cienegas was unchanged for a longperiod of time, maybe for more than a few millionyears. Not only cichlids and other fish are endemic tothis area but also snails, crustaceans, and reptiles.

“Cichlasoma” minckleyi has received a lot of at-

tention from evolutionary biologists as it is a poly-morphic species which could give some clues as tohow polymorphism could be associated with the for-mation of species (Kornfield et al., 1982; Liem &Kaufman, 1984). “C.” minckleyi is polymorphic inits shape and in the structure of the pharyngeal teeth.One morph is deep-bodied and has very large, molari-form teeth on the pharyngeal bones, whereas anothermorph has a cylindrical shape and sharp, papilliformpharyngeal teeth (see accompanying photos of at leastfour different morphs). Several investigations (Sagek. Selander, 1975; Kornfield et al., 1982) revealed thatother genetic characters, which would indicate thatwe are dealing with different species, are not coupledto the different morphs . These investigations alsoproved that the different morphs interbreed resultingin different morphs in one brood. One can thus con-clude that, even though the morphs look rather differ-ent, they still recognize each other as belonging to onespecies. After it became known that the CuatroCienegas cichlids represent just one, polymorphic,species and not four different ones (La Bounty, 1974),“C.” minckleyi was used in hypotheses as to how(sympatric) speciation in cichlids could take place ingeneral. One theory (Liem & Kaufman, 1984) hypoth-esises that the different morphs would one day mateassortatively and give rise to new species. At present,

72

The elongated morph of “Cichlasoma” minckleyi. This specimen was about 18 cm in size.

however, all morphs recognize each other as one spe-cies and the different morphs seem to feed less spe-cifically than one would expect from their differencesin pharyngeal morphology.

Liem & Kaufman (1984) found it unlikely that “C”.minckleyi originated from hybridisation between twospecies since the molariform morph is so well devel-oped as to point to a specialisation. But why hybridi-sation between two species? Couldn’t we explain itby the assumption that an early population of “C.”minckleyi became split up over several springs andadapted itself to the local circumstances? Some ofthese springs could temporarily have almost dried outand could thus have started a specialisation towardseating snails, an item which was predominantlyavoided by the ancestral population. Some othersprings could have had a particularly rich populationof small fishes on which the local cichlids could havespecialised. “C.” minckleyi seems to be a very vari-able species; in the almost-dried-out spring the indi-viduals with strong molariform pharyngeal teeth wouldhave an advantage over the generalised morphs,whereas in the fish-rich spring fast swimming indi-viduals would have an advantage. Both these morphscan eat any type of food, as is proven in the laboratoryand by field studies, but the highly specialised mo-

lariform morph indicates that there must have been astrong selection for this specialisation. We can hardlyexplain this highly advanced specialisation by the vari-ability of “C.” minckleyi alone. A non-assortativemating system, still present today, would have aver-aged out the genetic characters promoted by the envi-ronment.

None of the reports about the Cuatro Cienegascichlid mentions that, beside the trophic morphs, thereare also at least three colour morphs. The most com-mon in the spring I visited in May 1991 was light-blue, the second most common was yellow, and rarelydarker, blue-speckled individuals were observed. Theyellow and light-blue morphs both had deep-bodiedand cylindrically shaped individuals. The darker blueindividuals were small (approximately 5 to 10 cm)but seem to have different body shapes as well. Alongwith this apparently permanent coloration, many verydark coloured individuals were observed. However,when such a dark individual was frightened it quicklychanged to its basic color. I have made similar obser-vations for “C.” labridens in Rio Florido near CuidadValles. Although no in depth examination was made,it seemed that only elongated individuals of “C.”minckleyi turned black and that they did so when hunt-ing (small fish?).

73

A deep-bodied, light-blue morph “C.” minckleyi.

A deep-bodied, yellow morph “C.” minkleyi.

The hunting(?) colour is fading in this yellow individual.

An intermediate morph “C.”minckleyi.

If we combine the variability in trophic characterswith the presence of different colour morphs then weare likely to conclude that these characters have de-veloped independently in isolated populations. Therethey could have evolved into highly advanced formswithout the averaging effect of interbreeding. Onlyafter these isolated or even semi-isolated populationshad evolved into geographical variants, adapted to theirspecific environment, were they united by the build-ing of canals connecting the once isolated springs.Speciation had not advanced far enough to abolishspecies recognition among the different populationsand thus interbreeding produced the variable popula-tion as we know it today.

References

KORNFIELD, I. L., D.C. SMITH, P.S. GAGNON & J.N.TAYLOR (1982) The cichlid fish of Cuatro Cienegas,Mexico: Direct evidence of conspecificity amongdistinct morphs. Evolution 36, pp. 658-664.

LABOUNTY, J. F. (1974) Materials for the revision ofcichlids from northern Mexico and southern Texas,U.S.A.. Ph.D. thesis. Arizona State Univ.

LIEM, K. F., & L.S. KAUFMAN (1984) Intraspecificmacroevolution: functional biology of the polymor-phic cichlid species Cichlasoma minkleyi. in: Evo-lution of Fish Species Flocks (eds. AA. Echelle andI. Komfield). University of Maine at Omno Press,Orono, Maine, pp. 203-215.

MINCKLEY, W.L. (1969) Environments of the bolson ofCuatro Cienegas, Coahuila, Mexico, with specialreference to the aquatic biota. Texas Western Press,Univ. Texas, El Paso, Sci. Ser. 2; pp. 1-63.

SAGE, RD. & R.K. SELANDER (1975) Trophic radiationthrough polymorphism in cichlid fishes. Proc. Nat.Acad. Sci. USA. Vol. 72, No. 11, pp 4669-4673.

74

“Cichlasoma” septemfasciatum Regan, 1908Willem Heijns

The female of the Topaz Cichlid, “Cichlasoma” septemfasciatum, shows a brighter coloration than the male.

A male of the Topaz Cichlid. Photo by Willem Heijns.

Shortly after the creation of Central America by tec-tonic activity the narrow stretch of new land was com-pletely devoid of a freshwater fish fauna. But, as mightbe expected, this situation didn’t continue for long. Agreat colonisation spread from the south to occupythe newly created land. Cichlids probably played animportant role as they were among the fishes able tomigrate via the sea.

How exactly the colonisation took place remainsunknown, but the populations of fish entering the newland found themselves confronted with a continentaldivide, and they could colonize either the Pacific or

the Atlantic drainage. Once settled they could not crossthe divide, which acted as a barrier that kept the popu-lations on either side isolated. Populations derivedfrom one original species could thus develop inde-pendently, sometimes into new species.

This mechanism has led to the evolution of a numberof sibling species, especially in the southern part ofCentral America. One example is by “Cichlasoma”sajica and “C.” septemfasciatum, where the first isrestricted to the Pacific and the second to the Atlanticdrainage.

“C.” septemfasciatum is distributed over a muchlarger area than “C.” sajica. This has resulted in sev-eral populations which differ mainly in coloration.Many colour variants of “C.” septemfasciatum are thusknown. An overview of these races can be found in“Buntbarsche der neuen Welt: Mittelamerika” byWerner and Stawikowski.

A new population of this species was discovered in1986 in the Rio Sixaola, in the south of Costa Ricanear the border with Panama. This race of “C.”septemfasciatum became known as the Topaz Cichlid.It is characterized by a vibrant yellow-orange colora-tion which is also visible in males. Initial reports onthis cichlid mentioned that a formal description wason its way but this has never materialized.

75

Theraps coeruleus Stawikowski & Werner, 1987Willem Heijns

The female Theraps coeruleus is characterized by the blue spot in the dorsal fin. Photos by Willem Heijns.

The blue spot lacks in the male T. coeruleus.

In 1985, Stawikowski and Werner discovered a smallcichlid which they named “Small Blue” (Kleine Blaue)and brought it back alive to Germany. In the sameyear, after the authors had published a report abouttheir new discovery, Seeger and Staeck describedTheraps rheophilus and remarked that the newly dis-covered cichlid of Stawikowski and Werner could bea variant of their species. The “Small Blue” indeedresembled the specimens which were depicted in thedescription of T. rheophilus. Later it became clear thatSeegers and Staeck had been mistaken and T.rheophilus was designated a junior synonym of T.lentiginosus. In 1987 Stawikowski and Werner de-scribed T. coeruleus.

The type locality of T. coeruleus lies 30 km southof Palenque in the Rio Mizol Há Mexico. The river isa tributary of the Rio Tulija which, in its turn, is anaffluent of the Rio Grijalva. In the description of T.coeruleus Stawikowski and Werner provide some char-acteristics by which the genus Theraps can be distin-guished from other cichlasomini. This forms a basison which cichlids that comply with these characteris-tics can be placed in the genus. The type species ofthe genus is T. irregularis. Other species belonging tothe genus are T. lentiginosus and T. coeruleus.

All three species are rheophilic, i.e. they inhabit fast

flowing water. This also could be the reason why T.coeruleus was not been discovered earlier. One doesnot expect to find cichlids in the middle of fast flow-ing streams, and when one does they prove to be dif-ficult to catch.

T. coeruleus feeds on insect larvae which are foundamong the pebbles on the bottom of the stream. Suchsmall stones are turned over and the prey secured by aquick bite, preventing it from being carried away bythe stream. This behaviour is also observed in theaquarium, Apart from the crystal clear water they de-mand, T. coeruleus is an easy to maintain and veryattractive cichlid.

76

Thorichthys pasionis (Rivas, 1962)Willem Heijns

A male Thorichthys pasionis. Photo by Willem Heijns.

Since the revision of the genus Cichlasoma byKullander (1983) the nomen Thorichthys has becomemore and more current, in scientific as well as inaquaristic literature, for species which formed thatsection within the original large genus. The genusThorichthys comprises not more than eight species,five of which form a natural subgroup. T. aureus isthe best known of this group. The other three form anatural group as well, of which the Firemouth Cichlid,T. meeki, enjoys great popularity among aquarists. Theother two species of this subgroup are T. affinis and T.pasionis. The two subgroups differ in the shape of themouth. The lower jaw of T. pasionis and sister speciesprotrudes slightly further than the upper (prognathous)whereas the jaws in the species of the other subgroupare equally long (isognathous). The feeding techniqueof both groups may therefore differ as well; the prog-nathous species may scoop their food from thesubstrate whereas the isognathous cichlids pick theirchoice out of the sediment. However, nothing factualis known about their natural feeding behaviour.

T. meeki is the best known species of the genus. T.pasionis is closely related to this cichlid and the onlydifference, besides its distribution, is its yellow col-oration. This has led to its trade name of Yellow Meeki.Indeed, T. pasionis closely resembles the two other

species of the subgroup. The basic coloration of thebody is violet-blue; the head and lower part of the bodyshow many yellow areas. The most prominent char-acteristic, however, is the colour of the throat; darkbordeaux-red to almost black. The colour of the throatin T. meeki is red and in T. affinis is yellow. Speci-mens of T. pasionis in breeding coloration have a blackthroat. Normally the colour on the throat is not vis-ible, but this makes its effect more dramatic when thefish lowers the buccal cavity in a threatening display.The sharp contrast the suddenly appearing black throatforms with the yellow colour on the head must act asan effective means to deter intruders.

T. pasionis is found in the Rio Usumacinta drain-age (Rio de la Pasion) and in the Lago Petén Itzá inGuatemala and southern part of Mexico. T. pasionisis found in streams as well as in lakes. It forages oversometimes muddy bottoms searching for small preysuch as insect larvae and crustaceans.

In contrast to some other species of the genus, T.pasionis is relatively easy to keep in an aquarium. Theyquickly adjust to the conditions of the aquarium andtake any type of food offered. The only requirementthey have is for clean, oxygen-rich water.

77

SOUTH AMERICANCICHLIDS

Crenicichla species from the Rio XingúFrank Warzel

A female Crenicichla sp. “Xingú III” with typical coloration (TL approx. 20 cm). Photos by Frank Warzel.

For many, the Rio Xingú is a shallow 2000 km-long tributary of the Amazonas which is, by virtueof its many rapids, inaccessible to larger boats. Forichthyologists and interested aquarists the RioXingú is the supplier of an inexhaustible array ofunusual tetras, catfish, and cichlids. Recent inves-tigations have indicated that most fishes in the RioXingú system are endemic to these waters. A largepart of the fauna still awaits a scientific descrip-tion although publications concerning the region’sfish-fauna have increased recently. On the otherhand it seems that many species, which have al-

ready been imported for aquaristic purposes, areabsent from museum collections. This will certainlyhamper their description in the near future.

In the group of nameless cichlids we find sev-eral Crenicichla species of which the first was col-lected in 1988 near Altamira by Schliewen,Stawikowski, and Kilian. A few weeks after its dis-covery juveniles of this cichlid, introduced asCrenicichla sp. “Xingú”, were exported fromBelém.

At the same location a year earlier, in the þ sum-mer of 1987, Bergleiter hooked two large Creni-

78

1. Altamira; 2. Cachoeira do Espelho, type locality of C.percna; 3. Rio Fresco, type locality of C. phaiospilus; 4.Cachoeira von Martius, collecting site of C. phaiospilus; S.Suiá Missú, collection site of C. sp, “Suiá Missú”.

Crenicichla sp. “Xingu I”, juveniles (TL c. 15 cm).

Crenicichla sp. “Xingu I”, adult male.

Crenicichla sp. “Xingu I”, female in breeding coloration.

cichla species that appeared to be unknown to sci-ence. Reports that a pitch-black Crenicichla wasobserved near rocks and in large caves (Schliewen,pers. comm.) indicated that at least two species werepresent. Further imports from Altamira proved thatnot one but two other species were to be found there.These latter two were named C. sp. “Xingú II” andC. sp. “Xingú III” (Warzel, 1990a).

A fourth species from Altamira was exported inSeptember 1990 and introduced as C. sp. “XingúIV” (Warzel, 1990b). In the meantime this specieshas been scientifically described as C. percnaKullander, 1991. C. phaiospilus is known to occurin the Rio Xingú but has not yet been exported alive.A sixth species was discovered by Harald Schultzin the Suiá Missú, a tributary of the Rio Xingú.

It is likely that most (if not all) of these sixCrenicichla species are endemic to the Xingú sys-tem. One of the indicators of their supposed ende-mism is their relatively high developed specializa-tion. C. sp. “Xingú I”, for instance, was observedonly in shallow, fast-flowing sections of the river.It has a very slender body and in the aquarium itswims in a peculiar oblique position.

In contrast, C, sp. “Xingú III” has a, forCrenicichla, compressed body with a relativelylarge head and eyes. Observations of this speciesin the wild and also in part in the aquarium indi-cate that it prefers caves or dark recesses. C. percna,a species with an extremely shallow head and snout,is also highly specialized. Although reports aboutobservations in its natural environment have notbeen published, aquarium observations suggest thatwe are dealing with a bottom-oriented ambushhunter. I have observed that C. percna sometimesrests itself on a rock or on the bottom of theaquarium, almost feigning to be dead. The usuallylight-brown spots on its body then become very darkand form a conspicuous pattern. Such interestingbehaviour was frequently practised when smallerfishes could be seen by C. percna.

The specialization, if any, of C. sp. “Xingú II”and C. phaiospilus is at present unknown. Bothspecies have a shallow pointed head and high-po-sitioned eyes. C. sp. “Suiá Missú” from the upperXingú has a “normal” shape and is the only speciesof the six to have a head profile which resemblesthose of C. johanna, C. strigata, and C. lenticulata.Adults of latter species have a rather rounded snout.In contrast to the other five species, C. sp. “SuiáMissú” was not found in the vicinity of rapids orfast flowing water. Lowe-McConnell (1991) found

79

Crenicichla sp “Xingu I”, adult female.

Crenicichla sp “Xingu II”, adult female.

Crenicichla sp “Xingu III”, adult female.

Crenicichla percna, adult. Photos by Frank Warzel.

this species in a small lake-like widening of a streamnear Córrego do Gato.

C. sp. “Suiá Missú” has a characteristic colourpattern which is different from all other Creni-cichla. The ground colour is grey-green withround, black spots in the dorsal portion ofthe body. These spots are bigger andmerge together towards the end of thebody. Only males seem to have spotteddorsal fins. The red spots in the spinous part ofthe dorsal fin are unique in Crenicichla.

C. percna and C. phaiospilus also show a pecu-liar colour pattern. C. percna (Gr. perknos = darkspotted) has a leopard pattern with three or fourblack spots on its flanks whereas C. phaiospilus(Gr. phaios = dark, spilus = spot) shows four to fiveblack blotches on the side.

The pitch-black coloration of C. sp. “XingúIII” is also uncommon. Only one otherspecies in this genus is known to havesuch a coloration, namely C. cametanaSteindachner, 1911 from the Araguaiaand the Rio Tocantíns which are not farfrom the Rio Xingú. This does not automaticallymean that both species are closely related. C.percna, C. phaiospilus, and C. sp. “Suiá Missú” lackthe normally present light-bordered spots at the baseof the caudal peduncle. The colour patterns of C.sp. “Xingú I” and C. sp. “Xingú II” fall in the nor-mal range of Crenicichla, although both have ex-ceptionally intensive red hues, especially in theunpaired fins. Rarely, and then only in excited emo-tional phases, a zebra-like colour pattern is seenon C. sp. “Xingú I” consisting of about 9to 10 thin, vertical bars which reach half-way to the ventral region of the body. Outof all the Xingú species, it is mainly juve-niles of the latter species that have been imported.They have an intense orange-yellow ground colourwith a pattern of horizontal stripes which remainto a size of about 20 cm. Such a juvenile pattern isalso known for other species in this group. As longas these cichlids have this pattern, their behav-iour is rather peaceful and theyshow a tendency to formschools. It is somewhat moredifficult to mix C. sp. “XingúII” and C. sp. “Xingú III” withother fish. In particular juveniles of C.sp. “Xingú III” are pugnacious towards conspecificsand become solitary at an early age. Specimens ofC. sp. “Xingú II” first start to change their colour

80

Crenicichla sp. “Xingú II”, juvenile (c. 12 cm TL).

Crenicichla sp. “Xingu II”, female in breeding coloration.

Crenicichla phaiospilus (above) and Crenicichla sp. “Suiá Missú” (below). Drawings by Frank Warzel.

pattern at a size of 10 cm. C. sp. “Xingú III” juve-niles, however, at this size have a completely darkpattern and are indistinguishable from large adults.Even the typical iris-crescent behind the pupil isobvious in small specimens.

Although it may seem like that these six specieshave a rather distinct appearance and behaviour,they all have morphometric characters in commonwhich place them in a group apart from otherCrenicichla. If one is considering maintaining oneor several of these pike cichlids in the aquarium,attention should be paid to the individual sizes ofthese fish. Most grow to a length of more than 30cm, with the possible exception of C. percna, andtherefore need a large, at least 200 cm long,aquarium. In addition, these pikes grow enormouslyfast and juveniles of 10 cm may add 15 cm in thefirst year after importation. Large tanks with plentyof swimming space are a must, especially for C.sp. “Xingú I”. The chemical composition of thewater seems to be of minor importance in so far asthe well-being of these cichlids is concerned. Ow-ing to the fast turnover of food, frequent waterchanges are necessary and should keep the buildupof nitrates to a minimum.

Until now none of the Xingú species has been

81

Crenicichla percna, an adult specimen of unknown sex from Altamira. Photos by Frank Warzel.

spawned in the aquarium. An analysis of the waternear Altamira (Schliewen & Stawikowski, 1989)showed (September, 1988) a temperature of 32.2°C, a pH of 6.5, a hardness of circa 1° DH and aconductivity of 120 µS/cm. Even under perfectlyimitated water conditions, a spawning could not beinduced. Under these circumstances, however, thefemales showed their breeding coloration and hadripe ovaries. Females of C. sp. “Xingú I” attractedmales by approaching them in a bent, U-shapedposition of the body, with the central part of thebody became very light and conspicuous at the sametime. It is known from other Crenicichla that fe-males bend their bodies in an S-shape during thecourtship ritual. Hopefully more specimens will beexported in near future, so that we may increaseour knowledge of the behaviour and distribution ofthese impressive and beautiful pike cichlids.

References

KULLANDER, S.O. (1991) Crenicichla phaiospilus andC. percna, two new species of pike cichlids(Teleostei: Cichlidae) from the Rio Xingú, Brazil.Ichthyol. Explor. Freshwaters. Vol. 1, (4), pp 351-360.

LOWE-MCCONNELL, R.H. (1991) Natural history offishes in Araguaia and Xingú Amazonian tributar-ies, Serra do Roncador, Mato Grosso, Brazil.Ichthyol. Explor. Freshwaters. Vol. 2 (1), pp 63-82.

SCHLIEWEN, U & R. STAWIKOWSKI (1989) Teleocichla.DATZ (42) 8; pp 227-231.

WARZEL, F. (1989) Neu importiert: Crenicichla ausBrasilien. DATZ (42) 8; pp 456-457

WARZEL, F. (1990a) Neu importiert: Crenicichla ausdem nördlichen Brasilien (II). DATZ (43) 12; pp713-714.

WARZEL, F. (1990b) Neu importiert: Crenicichla spec.“Xingú IV”. DATZ (44) 1; p 7.

82

Crenicichla sp. cf. reganiFrank Warzel

A female Crenicichla sp. cf, regani from the Rio Xingu near Altamira, Brazil, in breeding coloration. Photos by Frank Warzel.

A male Crenicichla sp. cf. regani.

Before I report on Crenicichla sp. cf. regani, another,closely related, dwarf Pike Cichlid – C. regani – shouldbe mentioned. This attractive small Crenicichla wasexported in the early seventies and became knownunder trade names like “Nanus” or “Dorsocellata”.Crenicichla regani was described by Ploeg (1989)from a large collection of specimens. This dwarf PikeCichlid seems to have a wide distribution in Amazonia.Ploeg presumed that C. regani was not present in theRio Negro and that here its niche would be occupiedby C. notophthalmus. Windisch (pers. comm.), how-ever, found dwarf Pikes in the Rio Negro (togetherwith C. notophthalmus) which showed a close resem-

blance to C. regani. Underwater photographs, takenby Bergleiter (pers. comm.) in the lower Rio Xingunear Souzel, showed C. regani in its natural habitat.

The dwarf Pike depicted in the photographs,Crenicichla sp. cf. regani, was collected by RainerHarnoss in a small ditch near Altamira, Rio Xingudrainage, Brazil, about 200 km upstream from Souzel.This cichlid differs from C. regani in having a smalleradult size, a shorter snout, a deeper body, and con-spicuous markings in the tail. The maximum size ofC. sp. cf. regani seems to be about 6 cm for males and45 cm for females.

At a length of 4 cm the female assumed breedingcoloration. Instead of courting her “own” partner sheshowed off, with her body bent in the typical S-posi-tion and fins erect, to a much larger male C. regani inan adjacent aquarium. Unfortunately the female ig-nored her partner during the entire period during whichit seemed that she was willing to spawn. After aboutsix weeks she lost her beautiful coloration, which iscommon for Crenicichla. Unusually the conspicuousblack and white marking in the dorsal fin disappearedas well. This phenomenon is known only from PikeCichlids of the C. lugubris complex; they too can“switch off” the coloration in the dorsal fin when theystop breeding.

83

Acaronia vultuosa Kullander, 1989Frank Warzel

Acaronia vultuosa from the Rio Atabapo, Colombia. Photo by Frank Warzel.

Bocca de juquiá (giant-mouthed fish); this is the nameBrazilian fishermen gave cichlids of the genusAcaronia. They gained this name because of their large,oblique mouth with which they are able to suck insometimes rather large sized prey at an astonishingspeed.

Besides Acaronia nassa, the newly described A.vultuosa has also been infrequently imported as anaquarium fish. Both species are identified by theircharacteristic markings on the head. A. vultuosa has avery conspicuous pattern consisting of several lines.The pattern in A. nassa has broader bands and roundspots which in part have a light edge. The two spe-cies, which show a great resemblance towards eachother, have a mostly allopatric distribution and areregarded as sibling species. A. nassa is found in theAmazonas drainage while A. vultuosa lives in theOrinoco system. Only in the upper Rio Negro haveboth species been found sympatrically. This is prob-ably caused by the Casiquiare which forms a connec-tion between the two river systems.

In Spring 1991 we found A. vultuosa in the RioAtabapo, south of the Casiquiare. Its biotope, how-ever, was not in the river itself but about 100 m fromthe bank in a lagoon fed with water from a spring. Wefound many A. vultuosa in different stages of growth

mainly at the heavily vegetated edge of the lagoon.Habitat descriptions in the literature confirm that A.vultuosa prefers stagnant or slowly flowing water. Itspreference for this type of water is further indicatedby its typical swimming behaviour: it swims in shortbursts, stopping suddenly while scanning the area forpossible prey. Normally, however, A. vultuosa hoversamong the weeds and ambushes its prey.

Although Acaronia lives in soft, acidic water, itaccepts regular tap water without any signs of dis-tress. It takes any type of frozen food but ignores flakesor pelleted aquarium fare. A. vultuosa is not a cichlidfor a regular community aquarium. It can show manydefensive and threatening postures but in general itbehaves a little shyly towards other fishes, even to-wards cichlids that are noticeably smaller.

A. vultuosa remains smaller than A. nassa; the big-gest specimen recorded measured about 18 cm. Thelargest specimens we collected in the Rio Atabapowere still a few centimetres short of this length. Fur-ther observations concerning the behaviour of thiscichlid have not yet been published. One of these daysI will dim the lights of the tank in order to find out thereason for its large eyes.

84

Guianacara sp. “Red Cheek”Ron Bernhard

A pair of Guianacara sp. “Red Cheek” from Venezuela. Photo by Ron Bernhard.

In 1902, the French ichthyologist Pellegrin describedthe cichlid Acara geayi in honour of the collector F.Geay, who caught a number of specimens in the RioCamopi in French Guiana. Later the species was placedin the genus Aequidens and until 1989 it was knownas A. geayi (Pellegrin, 1902).

The popularity of this cichlid among aquaristsspawned a vast amount of publications which madethe name Aequidens geayi well-known amonghobbyists.

The Swedish ichthyologist Sven Kullander is knownfor two reasons. Firstly for his meticulous morpho-logical examinations of South American cichlids andsecondly for his ability to produce an infinite numberof descriptions of species and genera. As early as in1980 he indicated his doubts about the placement ofgeayi in Aequidens and pointed to the relationship thiscichlid has with those of the genus Acarichthys. Nineyears later Kullander not only introduced a new ge-nus but also two subgenera and three new species!The genus Guianacara and subgenera Guianacara andOelemaria are described in “The Cichlids of Surinam”coauthored by Han Nijssen. The species G. geayi, G.owroewefi and G. sphenozona belong to the subgenusGuianacara, whereas the subgenus Oelemaria con-tains only G. oelemariensis.

The cichlid which is depicted in the photographabove resembles G. owroewefi in many characteris-tics. The main feature of this species is the black spoton the anterior part of the dorsal fin and a black verti-cal stripe which is most intensively coloured on andjust below the upper lateral line.

Seen from a geographical standpoint the “RedCheek Geayi” is unlikely to be conspecific with G.owroewefi, as it is found in Venezuela whereas G.owroewefi is distributed over central and east Surinam.These two populations are separated by that of G.sphenozona which inhabits the river systems (mainlythe Corantijn River) in the western part of Surinam.

G. oelemariensis seems to be endemic in theOelemari River where it is sympatric with G.owroewefi. Without doubt the “Red Cheek Geayi”belongs to the subgenus Guianacara and will be re-ferred to as Guianacara sp. “Red Cheek”. The “RedCheek Geayi” is one of the most beautiful species ofthis genus. Its maximum length is about 14 cm formales and about 10 cm for females. It has been bredin captivity where it proved to be a cave spawner. Thefemale may disappear into the cave for the entire pe-riod during which the eggs develop into free swim-ming fry. The pair bond is strong and they continue todefend their territory when no fry are present.

85

Krobia speciesRon Bernhard

The red-eyed Krobia species was collected in French Guiana. Photos by Ron Bernhard.

Krobia sp. “Green”, a female fanning her eggs.

A male Krobia guianensis. Photo by Ad Konings.

Before Kullander’s revision in 1989 the genusAequidens contained many South American cichlidswhich could be placed in several groups. ThenKullander placed the species which were consideredas the guianensis-group in the new genus Krobia. Thebest-known species of the former group was knownto aquarists by the name itanyi, but later it turned outto be guianensis. These two species, now Krobiaguianensis and K. itanyi, differ from each other onlyin their breeding pigmentation patterns. The patternof K guianensis changes during breeding to verticalbars on the body whereas that of K. itanyi remains ahorizontal stripe.

When Kullander (1989) defined the new genus hementioned that besides the two described species inKrobia there were two others known to occur in FrenchGuiana. In the aquarium hobby two undescribed spe-cies of this genus are known, of which one, Krobiasp. “Red Eye” was collected in French Guiana. Theother species, K. sp. “Green” came from an unknownlocation and has already disappeared from theaquaristic scene. Krobia species are open substratespawners. The eggs are always laid on the same spotin the territory, which is defended by the pair, evenwhen no fry are present. The pair bond is strong andremains for years.

86

The “Aufwuchs-feeder” let the cichlids eat in a natural way and prolongs the feeding period.

The “Aufwuchs-feeder”Roger Häggström

When it comes to the mbuna of Lake Malawi, it iswell known that they spend most of their time feedingfrom the Aufwuchs on the surface of rocks. In a nor-mal community aquarium, though, the fish don’t spendmuch time feeding this way. They don’t spend muchtime feeding in any way. Ever since I have been keep-ing mbuna, I have wanted to give them the chance tofeed in a natural way. The result of my deliberationsis the “Aufwuchs-feeder” which I describe below.

This simple device consists of a PVC tube, a squareof plastic gauze, rubber bands, four plastic bars, andfour rust-free screws. The idea is to put the food be-tween the tube (rock equivalent) and the gauze so thatthe fish have to scrape the food through the gauze inorder to eat.

A PVC tube (20 cm long and with a diameter of 10cm) is cut lengthwise into two halves which can beused for two “Aufwuchs-feeders”. Drill a hole in oneend (the “top”) of the half-tube and tie a plastic stringin this hole. The string should be as long as your tank’sheight plus 10 cm. Tie a handle of plastic or wood atthe other end of the string. This handle should hangover the rim of the aquarium thus making it easy toremove the feeder from the tank.

The plastic gauze (18 x 15 cm) can be purchased insome hardware stores. It is used as insect-screen. It

should be a little stiff for easy maintenance. The meshshould not be more than 2 mm. Brass or stainless steelscreens work as well Cut four bars out of a sheet ofPVC with the sizes 20 x 1 cm and about 2 mm thick.Glue (PVC glue) the plastic net between two bars oneach side of the long end (see photo). These bars willpull the gauze tight over the half-tube. A hole is drilledin each end of both plastic (double) bars. The fourscrews are mounted in these holes (see photo). On oneside a rubber band is tied to each screw.

Now you can load the feeder. Use whatever food isavailable as long as it is not too hard or soluble in thewater. Pelleted food or frozen food is placed on thenet and rolled flat using the half-tube. Then take theloose ends of the rubber bands, draw them round thetube and mount them on the screws on the other endof the net. Rinse the whole feeder in running waterand put it in the aquarium. At first it will take sometime before the fish discover the food, but then youcan watch them feeding in their natural way. This maypromote the growth of the lips of cichlids likePlacidochromis milomo or Protomelas ornatus. Nowit takes much longer before all food has been eaten. Ithink this device could be regarded as the equivalentof the wheel in a hamster cage.

87

The platform is glued to one of the tank’s sides. Photo by Gerard Tijsseling.

Breeding Tropheus the natural wayGerard Tijsseling

Tropheus moorii is one of the favourite cichlids fromLake Tanganyika. It has been bred in captivity by thou-sands of aquarists. Some of them practise early re-moval of larvae from the female’s mouth in order tohave a higher survival rate of the entire spawn.

Among organisms like birds and animals, naturehas provided a process whereby the newly-born indi-vidual learns some important “tricks” from its motherwhich help it to survive, and which can play an im-portant role at a later stage of its life, e.g. when choos-ing a partner. This process is called imprinting andusually takes place a few hours or days after birth.

In my opinion fry of T. moorii and similar mouth-brooders from Lake Tanganyika are also imprintedduring the first few hours after they have been releasedfrom the female’s mouth. In my experience, youngwhich had been taken out of the mouth at a prematurestage grew up to be bad mouthbrooding females thatswallowed or neglected the first three to five spawns.Young which had been released naturally, grew up tobe excellent females that brooded the first spawn tomaturity.

In a community aquarium it is hardly possible tolet females release their babies naturally; in most casesthe fry will be eaten in minutes. Therefore I thoughtof imitating the situation in the lake where fry are re-

leased in the extreme shallows. These areas have manyadvantages; they are inaccessible to larger fish (preda-tors) and have plenty of food. Algae-eaters find mostof their food in the shallows as these are the lightestareas. Small rocks and pebbles provide a substrate forthe algae and the necessary shelter.

The photograph shows how a “shallows” can bebuilt onto your tank. Among my friends this setup isknown as “tank with balcony”. The platform is gluedto one of the tank’s sides. The “balcony” in the photo-graph has a size of 70 x 50 cm. In my experience 15cm is the maximum depth for such a platform becausedeeper “balconies” will be occupied by territorialmales.

The “balcony” is filled with small stones. Care mustbe taken to provide oxygen-rich water (place waterinlet over the platform) and lots of light (to stimulatealgae growth). The best thing is: it works perfectly!When a female is about to release her fry she swimsup to the “balcony” and stays a week or more with heroffspring before she returns to the group in the rest ofthe tank, The young, however, stay on the “balcony”till they have reached a size of about two centimetres.Not only Tropheus but also Goby cichlids (Eretmodini)use the “balcony” with success.

88

The cichlid bibleMartin Geerts

About twenty years ago C. M. Yonge (in his forewordto Fryer & Iles, The cichlid fishes of the Great Lakesof Africa, 1972) described the Great Lakes in EastAfrica as “a field laboratory of evolution”. AlthoughYonge is a malacologist (mollusc-expert) and LakeTanganyika harbours a great variety of snails, we mayassume that he was thinking of the cichlid speciesflocks when he put these words on paper. The EastAfrican cichlids have since then been in the limelightof aquarists, ichthyologists and evolutionary biologists.A few years later Dr. P. H. Greenwood (1974, Cichlidfishes of Lake Victoria, East Africa) remarked that theincreased attention given to these cichlids had nega-tively influenced the studies of other African fresh-water fishes and even of neotropical cichlids.

The present situation does not seem to be much dif-ferent from then as we read in “Cichlid Fishes; Be-haviour, Ecology and Evolution”. This book, “thecichlid bible”, was published recently by Chapman &Hall (London). The text is a compilation of papers byalmost twenty authors, all authorities in their field,edited by Miles Keenleyside. The result of their ef-forts is a must for every “cichlidologist” as it containsall the knowledge concerning cichlids gathered in thelast decades. Curiously, this knowledge is not reflectedin handbooks about evolution.

The first chapter of the “bible” treats the intra-familial relationships of cichlids and is written byMelanie Stiassny. Interestingly, given Greenwood’sremarks, Stiassny places the cichlids of the East Afri-can lakes in a group called “The Rest”. The lampro-logine cichlids are an exception but during a congressheld in The Hague (28-8-1991) she expressed herdoubts as to the correct place this group has in hercladogram.

That not everything in the “field laboratory of evo-lution” has its proper place may be illustrated by thefact that Ribbink (Chapter 2: 38) regards the mbunaas a monophyletic group, while Kornfield (Chapter 5:121) speaks of “....as currently defined, the Mbuna isa paraphyletic group”. Tony Ribbink treats the ecol-ogy of the cichlids in Africa’s Great Lakes with au-thority, even though the emphasis is put on Malawiancichlids. Yamaoka, in his treatment of the “Feedingrelationships” (Chapter 7), restores the imbalance byreferring mainly to Tanganyikan cichlids. Yamaoka’scontribution is of great interest especially the para-graph titled “How to coexist?”. This paragraph alsomakes clear that much research is still needed to elu-cidate the way East African cichlids influence eachothers life styles. Furthermore Yamaoka discusses thetechniques with which East African cichlids exploit

different sources of food.Cichlids have a well adapted feeding apparatus and

this aspect is discussed by Karl Liem in a separatechapter. Liem’s contribution describes the functionalmorphology of cichlids. Trophic radiation is com-monly regarded as the main cause of success in cich-lids, which is clearly why Liem puts emphasis on thetechnique with which cichlids collect and process food.His final words, however, show the great need of ad-ditional research: “In short, comparative functionalmorphology has demonstrated the shortcomings of ourcurrent and generally favoured interpretations ofcichlid diversifications, as monophyletic in origin anddriven by competition in faunistically closed basins.”

In general biology the species concept is still a hotpoint of discussion. Greenwood’s chapter concerningthis issue is therefore of great interest. With regard torecent investigations on “Cichlasoma” managuense,he points to the necessity of making a distinction be-tween ecophenotypic variation and polymorphism. Onpage 93 Greenwood expresses his disbelief that thepresence of predators, like the Nile perch or thetigerfish, would hamper speciation. Even the catas-trophe in Lake Victoria, caused by the introduction ofthe Nile perch, could not change his opinion.

The impact the Nile perch had on the species flockin Lake Victoria is further discussed by Barel, Ligtvoet,Goldschmidt, Witte, and Goudswaard. The authors donot restrict themselves to a description of this eco-logical disaster but give information about speciesidentification, ecology, and morphology as well. Froma taxonomic point of view the cichlids from Lake Vic-toria seem to provide a so far unsolvable problem, evenat the generic level. This situation is also reflected inthe treatment of the genus Haplochromis in the re-cently published CheckList of the Freshwater Fishesof Africa (CLOFFA) (part IV). In this checklist Green-wood’s revisions of the genus are not accepted. Ac-cording to Barel et al. species identification is a fun-damental problem for which modern taxonomic pro-cedures offer no solution (p. 268). The authors finddifferences in colour patterns among haplochrominesthat are morphologically identical. Sexual selection,therefore, could provide for a necessary reproductiveisolation.

Mating systems and sexual selection are treated inseparate chapters. George Barlow discusses matingsystems among cichlids. He believes that the evolu-tion of mating systems proceeded from monogamywith biparental care (substrate-guarding species) topolygamy with maternal care (mouthbrooding spe-cies). Barlow comments on many aspects of the dif-

89

The nest of Cyathopharynx furcifer, here the variant in Moliro Bay, varies in height according to the depth at which it is built.

ferent breeding techniques and does so with author-ity. Nevertheless a few mistakes have slipped into thischapter. Tilapia rendalli is not a biparental mouth-brooder as is stated on p. 180. This species is a substratespawner and that puts the words of Barlow in a differ-ent perspective. On the same page Tilapia mariae istreated as a mouthbrooder. Aequidens vittatus (=Bujurquina vittata) is not found in Surinam. Accord-ing to Kullander & Nijssen (The Cichlids of Surinam)Barlow’s reference relates to Krobia guianensis.Barlow’s remarks concerning the effects of predationmake it clear that fry and juveniles run the greatestrisk of being eaten. Parental care is therefore of ut-most importance and is treated by Keenleyside (Chap-ter 9). The author reviews investigations which havebeen performed in the field and this makes his contri-bution of great value to hobbyists.

Spawning and guarding of fry demand a coordina-tion between the parents. Mark Nelissen discussescommunication between cichlids. He distinguishesseveral forms of communication. The importance ofcommunication is considered in the chapter writtenby David Noakes, which deals with the ontogeny ofcichlid behavior. Kenneth McKaye comments onsexual selection (Chapter 12). Sexual selection is cur-rently regarded as the prime factor in speciation among

cichlids. McKaye questions whether even the nestbuilding of certain populations of sand-dwelling cich-lids could lead to speciation. He writes “Hence weconclude that bower (= nest. Ed.) form evolves pri-marily in response to female choice.” Ad Konings(pers. comm.) does not believe so. In his opinion theway a nest is built is mainly dependent on the depth atwhich it is made. At deeper levels the females see thenests from above which makes the height less impor-tant.

Rosemary Lowe-McConnell describes the distribu-tion and ecology of South American and Africanriverine cichlids. She gives us an authorative over-view of the situation. The last chapter concerns thebreeding of cichlids for food and is written by R. Pullin.

“Cichlid Fishes; Behaviour, Ecology and Evolu-tion”, the bible, is indispensable for every “cichlidolo-gist”. From the text in the book it is apparent that about50% of the species in “the field laboratory of evolu-tion” await formal description, while many genera arein need of a better diagnosis. Moreover nothing isknown about the relationships between many differ-ent groups of species. This may be the reason whycichlids, a prime example of speciation, are so poorlyrepresented in books about evolution.

90

The specific name of this cichlid, Ophthalmotilapia nasuta, is an adjective meaning “having a (distinctive) nose” and endsaccording to the gender of the generic name, which is feminine. Photo taken in Moliro Bay, Zaïre.

Cichlid classicsMary Bailey

Recent years have seen several major revisions ofcichlid genera, with, in most cases, the original genusbeing restricted to a few of its former species, andnew genera being erected for the remainder. Frequentlythe gender of the new generic names has differed fromthat of the original genus, necessitating alterations tosome of the specific names as well. As Latin is nolonger generally taught in schools the average aquaristhas been left in a state of some confusion, and it ishoped that this article will cast a little light upon thesubject.

The language of taxonomy is nominally Latin, butmany of the word used are latinised forms of wordsfrom other languages (mainly Greek) and propernames from all over the world. The grammar of tax-onomy is, however, strictly Latin. Under the trinomialsystem of nomenclature, devised by Linnaeus in the18th century, all living beings are classified by threenames (the meaning of trinominal): the generic, thespecific, and the subspecific. All generic names arenouns (words which name a thing); “fish” is a nounwhich tells us with what part of the animal kingdomwe are dealing, and likewise Tropheus, Etroplus,Aequidens, are Latin nouns, generic names whichspecify particular groups of fish species.

Latin nouns all have a gender – masculine, femi-

nine, or neuter – which, in classical Latin, is usuallyrecognisable by the ending of the word. The bestknown are “-us” (masculine), “-a” (feminine), and “-um” (neuter). Unfortunately in taxonomy it is not quitethat simple, as if a generic name is a latinised form ofa word from another language, it retains its originalgender (if any). Thus Cichlasoma, which has Greek“roots”, appears feminine at first glance because ofits “a” ending, but is in fact neuter. It may be neces-sary to refer back to the original generic descriptionin order to establish the derivation, and hence the gen-der, of a generic name. Aquarists who find themselvesconfused may take heart from the fact that even thegreatest taxonomists can themselves sometimes be inerror – the Greek word “gramma” can mean “letter”,in which case it is neuter, or “line”, when it is femi-nine; Dr. C. T. Regan confused the genders of the twowhen erecting Apistogramma, and, while meaning“line”, used the neuter gender. This error remaineduncorrected until Kullander (1980) revised the genus.

If we wish to refer to a particular member of a ge-nus then we must use the specific name, eg Tropheusduboisi. The specific name “qualifies” the generic;“duboisi” tells us that we are dealing with a particulartype of Tropheus, that named after a M. Dubois. Itenables us to refer to this species in the literature, and,

91

Tanganicodus irsacae is named after an organisation. Photo taken near Kavalla Island, Zaïre.

to many of us, conjures up an immediate visual imagederived from our own data banks. Subspecific namesallow us to identify distinct genetic groups within aspecies when such exist; they follow the same gram-matical rules as specific names. The specific name canbe one of three types: an adjective (a descriptive word,eg “auratus”, meaning “golden”); a genitive (the pos-sessive of a name, eg “duboisi”, which means “ofDubois”); or a noun in apposition (a noun used toqualify another noun, eg “Mr Bun the Baker”, or,aquatically, Pseudotropheus zebra, “Cichlasoma”centrarchus).

The commonest type of Latin adjective has the “-us”, “-a”, “-um” endings (respectively masculine,feminine, neuter) mentioned earlier in respect of nouns;there is another common type with the endings “-is”(masculine and feminine) and “-e” (neuter), egApistogramma pertensis, “Cichlasoma” nicara-guense. It is worth noting in passing that in Latin ALLsyllables are pronounced, and that the correctpronounciation of the “ense” ending is “en-say” (twosyllables)!

Adjectival specific names MUST agree in genderwith the generic name to which they are applied, andso if the species is moved from one genus to another,with an attendant change of gender, the specific name

must be altered. Thus Cichlasoma severum (neuter) isnow Heros severus (masculine).

A genitive is the possessive form of a Latin noun,and is equivalent to “of X” in English. In taxonomy itis almost always used to name a species after a per-son, although there are rare exceptions – Tanganicodusirsacae is named for the organisation IRSAC, andTangachromis dhanisi after a boat, the “Baron vonDhanis”. Although as a rule adjectives are used to de-note place names, occasionally the genitive of such aname is used, usually where it is impossible to form asensible adjective, as with African names such asNkhata and Nkamba.

Genitives take the gender of their originator, withthe masculine ending “-i” (”regani”, “burtoni”), andfeminine “-ae” (”kingsleyae”, “trewavasae”). Thefeminine ending has also been used in the place namederivatives “nkatae” and ’hkambae” for reasons ofcommon sense and pronounciation*.

Very very occasionally a species is named after twopeople, thus we have “Cichlasoma” hogaboomorum,named for the Hogaboom brothers, “-orum” being themasculine plural ending. The feminine equivalent is“-arum”, but I can think of no example of its use.

Genitives are thus quite distinctive, and unlikely tobe confused with adjectives. They remain constant

92

whatever the gender of the generic name.Unfortunately nouns in apposition often have “-us”,

“-a”, “-um” endings like some adjectives, and whenthe species involved are moved to a different genus itis all too common for these names to be subjected to’blanket” gender changing by aquarists. But becausethey are nouns in their own right they have their owngender, which always remains the same whatever thegeneric name. Some are easy to spot, especially wherethey are the names of other animals – the derivationof Pseudotropheus zebra is obvious to everyone; ifyou know that Centrarchus is the name of anothergenus which ’Cichlasoma” centrarchus resemblesthen there is no problem, But there are also “ordinary”nouns used in apposition, and, to exacerbate matters,some of these are themselves composites of a nounand adjective. “Melanotheron” (”black-chin”) and“curviceps” (”curved-head”) will probably be left wellalone, but the same cannot be said for “maculicauda”(”spotted-tail”) or “longimanus” (”long-hand” [-fin])which look like adjectives but are not. Life becomeseven more confusing when one realises that there is aperfectly good adjective “maculicaudatus, -a, -um”(”having a spotted tail”) which must agree with anygeneric name to which it is applied.

It must be quite obvious, from the above, that with-out a lot of background knowledge, either of Latinand Greek vocabulary, or of taxonomic literature, theaverage aquarist cannot hope correctly to revise everyspecific name where generic change has occurred. Inconsequence it is wise to conduct a little research be-fore committing oneself to paper or in public.

Recently some authors have adopted the practiceof explaining the derivations and genders of the namesthey create; many earlier authors, however, omit toprovide any derivation, and gender/grammar are noteven mentioned as these used to be a matter of com-mon knowledge amongst scientists, who, until com-paratively recently, would automatically have receiveda “classical education”. In the absence of any clues inthe original description it can be helpful to examinethe subsequent synonymy of a species, as any earliervariation in the specific name will usually point to anadjective. The .”Cichlid Catalogue” of Ufermann,Allgayer, and Geerts (1987) can be extremely helpfulfor this purpose. Doubtless many aquarists will ques-tion whether there is any point in going to all this trou-ble.

Well, the point of any language, taxonomic or oth-erwise, is to make oneself understood, and the trino-mial system is intended to facilitate the communica-tion of species identity between people of different

languages. If it is to fulfil this purpose then it needs tobe used accurately and precisely. And, in an era wherescientists are beginning to take notice of observationsmade by aquarists, it is important for our credibilitythat we too are precise and accurate if we wish to makea meaningful contribution to the knowledge of ourcichlids.

References

KULLANDER, S.O. (1980) A taxonomical study of thegenus Apistogramma Regan, with a revision of Bra-zilian and Peruvian species (Teleostei: Percoidei:Cichlidae). Bonn. Zool. Monog. 14: 22.

UFERMANN, A., R. ALLGAYER, AND M. GEERTS (1987)Cichlid Catalogue. Brumath, France.

* For many years there has been confusion over thegenitive form of the name Moore (referring to J.E.S.Moore, collector of the type species of the genus Tro-pheus). Boulenger’s original description uses moorii,being the genitive of the latinised form “Moorius” ofMoore’s name; some authors maintain that “moorei”,a simple genitive of the English form, should be used.If, however, we refer to the rules of zoological no-menclature we find that there is no room for dispute.Any author describing a new species can choosewhether he latinises a name before using it, or simplyapplies the appropriate ending. It is not for subsequentauthors to question his decision. The name publishedin the original description of a taxon is the correct one(provided, of course, that there is no question of syn-onymy, preoccupation, or gender change). A name canbe altered only if it was misspellt by the typist or printer(but NOT if the mistake was the author’s) or if it isgrammatically incorrect. None of these apply inBoulenger’s case, so his published name Tropheusmoorii is the correct one.

93

A pair of Copadichromis eucinostomus (Msuli Point, 16 m) displaying the TB-position; see text for explanation.

Spawning techniques in mouthbrooders1. Fertilization in mouthbrooding cichlids

Dr. Ethelwynn Trewavas

Wickler (1962a,b) put forward two theories to explainthe evolution of the egg-mimicking spots on the anal finof males of many mouth-brooding haplochromines: 1.that they served to trigger spawning in the female, and 2.that the female, who has taken the eggs into her mouthvery rapidly as they are laid, is tricked into trying to pickup the egg-dummies and so to inhale sperm from themale’s vent and to ensure the oral fertilization of anyeggs not already fertilized. These theories have beenwidely accepted, but recently they have been challengedby Eva Hert (1989). Hert’s experiments showed that thespots in A. elegans attract and stimulate the female, al-though only if accompanied by courting movements ofthe male. The fertilization rate, however, was not sig-nificantly different between males with egg-spots andthose from which the spots had been removed by freeze-branding, both achieving nearly 100%.

Comparison of experimental with natural conditions

The experimental conditions were: one male andfive females were together isolated in each of 12 sepa-rate compartments. There were thus no rival males,no predators, no egg-robbers, no water currents. Theseare ideal conditions for undisturbed spawning. In con-trast abundant evidence exists that in nature a spawn-ing pair may be surrounded by any or all of these haz-ards (see McKaye, 1983; Ribbink et al. 1983;Konings, 1989: 291; 1990: 150, 204).

Adaptation to mouthbrooding

Mouthbrooding has evolved several times in the fam-

ily Cichlidae, in its most advanced form in the tribesTilapiini and Haplochromini (Peters & Berns, 1978 and1979). The tilapiines include both substrate- and mouth-brooders. Evolution from the former to the latter involvedthe regression and loss of features adaptive to substrate-spawning and guarding, and the development of newfeatures appropriate to the care of larger and fewer,nonadhesive eggs. The problem was, how to get the eggsrapidly into the maternal mouth without neglecting theirfertilization. Meeting this need there evolved a behav-ioural trait, first described by Baerends & Baerends-VanRoon (1950: fig. 50, 51) and named “tail-chasing” (Fryer,1959), T-positions (Heinrich, 1967), making half-loops(Balon, 1977) or “tight circles” (Kapralski, 1990b) witheach other. The alternating T-positions are: TA: male’ssnout contacts female’s body near the vent, TB: maleplaces his body before the female with his vent near hermouth. The series of T-positions culminates after a veryshort time or up to 30 min. (Berns et al., 1978) in the“real spawning”, when at TA the female lays an egg or abatch of eggs and at TB she collects them, either allow-ing the male to fertilize them first or more rapidly so thatshe has to inhale the sperm to fertilize them in her mouth.

It is this last movement, oral fertilization, that Wicklersuggested was prompted by the deception of the egg-mimics on the male’s anal fin in species possessing them.

The series of T-positions (TA-TB-TA-TB) is repeateduntil the female has no more eggs to lay. The importanceof circling with T-positions is that by mutual stimulationit synchronises the spawning acts of male and femaleand facilitates fertilization either immediately before,during or after the collection of the eggs (Mrowka, 1987).

Wickler’s prime exemplar of oral fertilization was

94

Astatotilapia (formerly Haplochromis) burtoni. His filmof this species shows the female making over the “egg-dummies” the same lip-movements that, immediatelybefore, she has used to suck in the real eggs. Descrip-tions of spawning in mbuna by careful aquarists give thesame picture, e.g. Mary Bailey (letter of 28-1-1989) de-scribes the apparent efforts of Pseudotropheus socolofito suck in the eggsspots as “persistent”. Wickler usedthe word “energisch” for the same behaviour in his “H.wingatii” (the species Wickler studied was Astatotilapiacf. bloyetii from Tanzania).

The female “H. wingatii”, however, sometimes givesthe male time to fertilize the eggs before she collectsthem. In that case she “does not deem the spots worth aglance” (Wickler, trans. E.T.). This indicates that whenshe does take an interest in the fin her behaviour is di-rected to fertilization rather than to egg-safety. Wickler’stheory, on the other hand, distinguishes between the fe-male’s probable motive (safety) and the survival-valueof her action (fertilization).

It is evident that “H. wingatii” is flexible in its tech-nique, as, in response to environmental situations,some other species are found to be (Konings, this vol.p. 97). A. burtoni seems to be more rigid in its use oforal fertilization. Paulo’s experiment (1975) with thisspecies, in which he removed a male’s anal fin, clearlyshows that the egg spots, and even the fin itself, arenot necessary for oral fertilization. This cannot be de-duced from Hert’s results because we do not knowwhether A. elegans used oral fertilization either dur-ing the experiment or otherwise.

The use of similar lip-movements over eggs and“egg-dummies” which seemed to support Wickler’stheory, is weakened as evidence when we realise thatthe same method, namely suction, would be used fortaking in either eggs or sperm. At this time the fin isprobably covered with sperm.

Occurrence of egg-spots

Tilapiine mouthbrooders and some haplochromineshave no bright yellow spots on the anal fin of males.Many Malawian species have a conspicuous pattern,usually yellow, that does not resemble eggs, but in mbunaand some others the orange to orange-yellow spots aresurrounded by a contrasting border or a dark fin and havebeen called egg-spots. The closest resemblance betweenspots and eggs characterises the haplochromines of theLakes Victoria, Edward, and Kivu (VEK) group and re-lated species of streams, lagoons and the inshore watersof lakes, e.g. A. burtoni and A. calliptera. In these thespots are of the same order of size as the eggs and re-

semble them in colour and shape (Goldschmidt, 1991),The contrasting border is well-marked and regular (orthe fin itself is dark). Their mimicry has been acceptedeven by some who reject Wickler’s explanation of it.Among the mbuna there are some with a pattern indis-tinguishable from this type, others in which the spots aremuch smaller and arranged in a cluster instead of in rows.Occasionally they may occur also on the dorsal fin or infemales. There is interspecific variation as well as a de-gree of uniformity within a species or a population(Konings, 1990 and pers. comm.). This gives some sup-port to the theory of Axelrod (see Ribbink in Jackson &Ribbink, 1975) that the female recognizes the species-specific pattern. This theory and that of Konings (1989,p. 29) that the displayed anal fin, especially when bear-ing a brightly coloured pattern, “shows the female whereto spawn” are applicable both to patterns not resemblingeggs and to egg-spots. Neither claims any influence ofthe spots on oral fertilization.

Oral fertilization without egg-spots

For haplochromines see Konings, this vol. p. 97. Fer-tilization in maternal mouthbrooding tilapiines has beendescribed for seven species among the three subgeneraof Oreochromis. These accounts are summarized inTrewavas, 1983 (pp. 190, 367, 387, 419, 450, 475, 487),where references to the original records are supplied. Insix species oral fertilization is inferred from the snap-ping actions of the female close to the genital papilla ofthe male during the laying and collecting of the eggs.The seventh species, O. (Nyasalapia) karomo, was ob-served in the Malagarazi Swamps, where male and fe-male glided successively over the nest evidently layingand fertilizing the eggs. Similar behaviour by O.macrochir was seen in the Lufira River, but in the labo-ratory the female mouthed the genital tassel of the male.Wickler suggested that the tassel might function as a kindof spermatophore. If so, it would be by retaining the sper-matic fluid (slightly more viscous than water) at the sur-face of its branches long enough for the female to obtaina concentration of sperm. The male’s genital pore is atthe base of the bifid tassel (see Trewavas, 1983, fig. 171).

The tassel in all species of this subgenus (Nyasalapia)is cream- to orange-coloured in contrast to the usuallyblack breeding colour of the male. The black male of O.mossambicus has a prominent yellow papilla. In O.(Alcolapia) alcalica grahami the white papilla is sur-rounded by a black ring, which is highly conspicuous onthe white skin of the belly as shown in a photograph byDr. Wickler in Albrecht et al. (1968), also reproduced inTrewavas, 1983. As Fryer & Iles (1972) note, conspicu-

95

Astatotilapia calliptera, Thumbi East, L. Malawi. The perfect egg-mimics. Note their position. See also photos on pp 54-57.

Aulonocara sp. “Chitande Type Masinje”. A lavish displayon a black fin.

Petrochromis sp., Bulu Point, L. Tanganyika. Egg-spots onboth anal and dorsal fins. Photos by Ad Konings.

ousness is an attribute shared by these features and thedecorated anal fin of haplochromines. See alsoGoldschmidt (1991), who seems to rate it as at least asimportant as the mimetic nature of the egg-spots.

As food for thought on this subject I call attentionto two published photos. One, by Spreinat (1990: 529),shows a pair of Rhamphochromis sp. in an early stageof courtship. The body and fins are of uniform colour,except the male’s anal fin, which bears a conspicuous

yellow pattern. The female is targeting this fin. Thesecond is a photo, by Milner, on p. 4 of Jackson &Ribbink (1975), of a group of partial albino Pseudo-tropheus zebra. These have no dark pigment – eventhe retina shows pink through the pupil – but in threeindividuals the anal fin bears a cluster of 2-4 smallorange-yellow spots in a black field. Nothing couldbe more conspicuous or less egg-like. What geneticquirk has here made an exception in favour of the mostvaluable asset of a male cichlid?

Summary and conclusions

In maternal mouthbrooding cichlids fertilization, es-pecially oral fertilization, is facilitated by variousmeans. Mutual circling with T-positions ensures in-semination either during or after collection of the eggs.The female is guided to the location of the male’s geni-tal papilla by a conspicuous device, – the colour ofthe papilla in contrast with its surroundings, an out-growth of the papilla (rosette or tassel) or a conspicu-ous pattern on the male’s anal fin. It is suggested thatthe tassel or fin delays the dispersion of the somewhatviscous seminal fluid and that it is for this that thefemale mouths tassel or fin at the end of a spawningbout if she still holds unfertilized eggs in her mouth.

This theory of the female’s interest in the male’s findoes not differentiate between the egg-pattern and anyother, and causes us to look again at Wickler’s secondtheory. An emergency that causes the female to pick upthe eggs unfertilized may also cause her concern for egg-safety to amount to an obsession and make her moreeasily deceived by the egg-dummies. Some studies offertilization already show that behaviour in this functionmay be influenced by the environmental situation (seeKapralski, 1990a and c) and more such studies are desir-able, using species with different anal-fin patterns. Theegg-pattern appears to be very successful, since it oc-curs in all species of Astatotilapia and related generaconstituting the VEK group of haplochromines as wellas several species of the more diverse flock of LakeMalawi and some of Lake Tanganyika. Wickler’s twotheories are the only ones that offer an explanation of itsevolution and they have so far not been disproved. Un-less at least one of them is accepted, then I doubt if weare justified in giving the resemblance the name of eggmimicry. After all, it is a very simple shape and the darkborder that gives the spots a three-dimensional appear-ance also makes them very conspicuous, a valuable as-set. The theory of mimicry must rest on the suppositionthat the egg-image has a special significance for thegravid female, and this has not been disproved.

96

Pseudotropheus tropheops OB morph (Chinyamwezi Reef).The TA-position: the male’s snout contacts the female’s bodynear the vent while she deposits some eggs.

Cost-benefit estimate

It has been objected (Konings, 1989: 30) that theconspicuously decorated anal fin may carry the dis-advantage of attracting predators, especially when thespots are egg-mimics. This is a special case of a di-lemma very common throughout the vertebrates. Evo-lution seems to have decided that the benefit of suc-cessful breeding outweighs the possible cost. This iswhy both we and the peahen can enjoy the display ofthe peacock, and why both female cichlids and fishphotographers are seduced by the brilliant colours ofmale cichlids.

Acknowledgments

I have benefitted by discussions with Dr. TijsGoldschmidt and with aquarist friends, especiallyMary Bailey and Ad Konings, as well as from theirpublished works. I am grateful to my friend Mrs. IsabelRampton for kindly typing the script.

References

ALBRECHT, H., R. APFELBACH & W. WICKLER (1968) Über dieEigenständigkeit der Art Tilapia grahami Boulenger, ihrenGrubenbau und die Zucht in reinem Sueßwasser (Pisces,Cichlidae) Senckenbeq,. biol. 49; pp 107-118.

BAERENDS, G.P. & J.M. BAERENDS-VAN ROON (1950) An intro-duction to the study of the ethology of cichlid fishes. Be-haviour suppl. I. 243 pp.

BALON, E.K. (1977) Early ontogeny of Labeotropheus Ahl,1927 (mbuna, Cichlidae, Lake Malawi) with a discussionon advanced protective systems in fish reproduction anddevelopment. Env. Biol. Fish. 2,”(2); pp 147-176.

BERNS, S., E.H. CHAVE & H.M. PETERS (1978) On the biologyof Tilapia squamipinnis (Günther) from Lake Malawi(Teleostei: Cichlidae). Amh. Hydnbiol. 84; pp 218-246.

FRYER, G. & T.D. ILES (1972) The Cichlid Fishes of the GreatLakes of Africa. Oliver and Boyd, Edinburgh & London.

GOLDSCHMIDT, T. (1991) Egg mimics in haplochromine cich-lids from Lake Victoria. Ethology. 88; pp 177-190.

GOLDSCHMIDT, T. & J. DE VISSER (1990) On the possible role ofegg-mimics in speciation. Acta Bioth. 38; pp 125-134.

HEINRICH, W. (1967) Untersuchungen zum sexualverhaltungder Gattung Tilapia (Cichlidae, Teleostei) und beiArtbastarden. Z. Tierpsychol. 24; pp 684-754.

HERT, E. (1989) The function of egg-spots in an African mouth-brooding cichlid fish. Anim. Behav. 37; pp 726-732.

JACKSON, P.B.N. & A.J. RIBBINK (1975) Mbuna. TFH Publ.,New Jersey. (Illustr, in colour) 128 pp.

KAPRALSKI, A. (1990a) Breeding Pseudotropheus tropheops.TFH June. pp 86, 89-92.

KAPRALSKI, A. (1990b) Spawning Labeotropheus fuelleborni.TFH July. pp 10-13, 17 illustr.

KAPRALSKI, A. (1990c) Breeding Haplochromis compressiceps.TFH August. pp 32-34, 37, 41.

KONINGS, A. (1989) Lake Malawi cichlids in their naturalhabitat. Verduijn Cichlids & Lake Fish Movies. Nether-lands. 313 pp. illustr.

KONINGS, A. (1990) Cichlids and all the other fishes of LakeMalawi. TFH Publ. Neptune City, NJ. 495 pp. illustr.

MCKAYE, K.R. (1983) Ecology and breeding behaviour of acichlid fish, Cyrtocara eucinostomus, on a large lek in LakeMalawi, Africa. Env. Biol. Fish. 8; pp 81-98.

MROWKA, W. (1987) Oral fertilization in a mouth-broodingcichlid fish. Ethology. 74; pp 293-296.

PAULO, J. (1975) Bedeutung der Eiflecke. DCG-Info (Ger-man Cichl. Ass.) 6; pp 163-165.

PETERS, H.M. & S. BERNS (1978) Über die Vorgeschichte dermaulbrütenden Cichliden. 1. Was uns die Haftorganen derLarven lehren, pp 211-217. 2. Zwei Typen von Maul-brütern. pp 324331. (illstr. in col.). Aquarium Magazin

PETERS, H.M. & S. BERNS (1979) Regression und Progressionin der Evolution maulbrütender Cichliden. Mittl. Hamb.zool. Mus. Inst. 76; pp 506-508.

RIBBINK, AJ., B.A. MARSH, A.C. MARSH, A.C. RIBBINK, AND

B.J. SHARP (1983) A preliminary survey of the cichlid fishesof rocky habitats in Lake Malawi. S. Afr. J. Zool 18 (3); pp149-310.

SPREINAT, A. (1990) Beobachtungen an Rhamphochromisarten.DATZ 43; pp. 528-533.

TREWAVAS, E. (1983) Tilapiine fishes of the genera Santher-odon, Oreochromis and Danakilia. 583p., London, Br. Mus.Nat. Hist.

WICKLER, W. (1962a) Zur Stammesgeschichte funktionellkorrelierter Organ- und Verhaltensmerkmale: Ei-Attrappenund Maulbrüten bei afrikanischen Cichliden. Z. Tier-psychol. 19; pp 129-164.

WICKLER, W. (1962b) Egg-dummies as natural releasers inmouth-breeding cichlids. Nature 194; pp 1092-1093.

WICKLER, W. (1966) Über die biologischen Bedeutung desGenital-Anhangs der männlichen Tilapia macrochir.Senckenbeq, biol. 47; pp 419-422.

97

A pair in TB-position. Note the egg near the anal fin. The “total oral fertilization” method is practised by C. borleyi.

2. The spawning-techniques of Malawian haplochromines

Ad Konings

All the haplochromine species of Lake Malawi arematernal mouthbrooders. Usually spawning beginswith the manoeuvre known as circling with alternat-ing T-positions (TA-TB: see Trewavas, this vol. p. 93)which brings the pair to a synchronised spawning actwhen the eggs are laid, fertilized, and taken into themouth almost simultaneously, While the female de-posits the egg(s) the male has its head close to thefemale’s vent (TA-position). This position is very char-acteristic for most mouthbrooders. All haplochrominemouthbrooders display the TB-position where the fe-male collects the eggs and/or sperm. As Wickler(1962a) recorded for his “H. wingatii”, some speciespractise fertilization either before or after the femalehas picked up the eggs. Recent descriptions byKapralski (1990) show that in some species behav-iour is flexible in response to the environmental situa-tion. I have observed that females collect the eggs morerapidly in a crowded situation than in a peaceful one.

The fertilization of the eggs outside the female’smouth (external fertilization) has now been recordedfor several Malawian species, e.g. Dimidiochromiscompressiceps, Pseudotropheus sp. “TropheopsChilumba” (Kapralski, 1990c, 1990a), Nimbochromislivingstonii, Cyriocara moorii, and Labeotropheustrewavasae (Konings, 1989). Most of these speciesare known to fertilize (partly) the eggs in the mouthas well, i.e. D. compressiceps (Knabe, pers. comm.),Labeotropheus spp. (Balon, 1977; Bailey, pers.comm.), C. moorii (De Langhe, Pers. comm.), Ps.tropheops (pers. obs.). It seems that these species prac-tise an oral fertilization when they feel disturbed. Wemay call this technique the “facultative oral fertiliza-tion”. It is not known whether these conditions arenatural or induced by the confines of the aquarium.Many more species may practise an environmentallyinduced variation of fertilization.

So far, N. livingstonii is the only species of the groupof which oral fertilization has not (yet) been observed.

It is therefore not enough to say “spawning by the usualhaplochromine method”. Moreover a second spawn-ing variant (more advanced?), whereby the TA-posi-tion is not practised, has been described (see Konings,1989).

This second technique is employed by (probably) alarge group of non-mbuna. I have observed it in C.borteyi, Maravichromis lateristriga, Protomelastaeniolatus and Copad. sp. “Kawanga”. During theentire spawning act the male never has its head nearthe female’s vent (TA-position). In order to bring thepair to a synchronized spawning act the male leadsthe female to the spawning-site where an initial TB-position culminates in a circling around each other.After a few rounds the male discontinues the circling.While the female discharges her eggs the male waitsbeside her (passively?) until she has collected the eggsin her mouth. Then he circles her again and quivershis anal fin over the nest. Although quivering, the maledoes not move forward but probably exudes its se-men. The female snaps at the male’s vent and prob-ably inhales the milt to fertilize the eggs inside hermouth. We may call this spawning technique the “to-tal oral fertilization” method as all eggs are fertilizedinside the female’s mouth.

Of the three fertilization techniques mentioned thefacultative oral fertilization method is the one mostcommonly observed in the aquarium. Facultative andtotal oral fertilization have both been observed in thenatural environment (pers. obs.: Ps. zebra and Ps. sp.“Elongatus Ornatus”, and Copad. sp. “Kawanga” re-spectively). For references see page 96.

Acknowledgements

I gratefully acknowledge the discussions I enjoyedwith Dr. Ethelwynn Trewavas about many aspects ofcichlid-ichthyology and with Gerard Tijsseling andPeter Baasch about spawning behaviour in cichlids.

98

AustraliaThe New South Wales Cichlid Society

P.O. Box 163Moorebank, N.S.W. 2170

Queensland Cichlid GroupP.O. Box 163

Wooloongabba, Queensland 4102

Victorian Cichlid Society23 Mangana Drive

Mulgrave, Victoria 3170

AustriaDeutsche Cichliden GesellschaftVictor Kaplan Straße 1-9/1/3/12

A-1220 Wien

BelgiumBelgische Cichliden Vereninging

Kievitlaan 23B-2228 Ranst

CzechoslovakiaSZCH Klub Chovatelov Cichlíd

Príkopova 2CS-831 03 Bratislava

DenmarkDansk Cichlide Selskab

Ornevej 58, st. tv.DK-2400 Kobenhavn NV

FranceAssociation France Cichlid

15 Rue des HirondellesF-67350 Dauendorf

GermanyDeutsche Cichliden Gesellschaft

Eberescheweg 41D(W)-4200 Oberhausen

HungaryHungarian Cichlid AssociationLukács Laszló, Karolina ut 65

H-1113 Budapest

NetherlandsNederlandse Cichliden Vereniging

Boeier 31NL-1625 CJ Hoorn

SwedenNordiska CiklidSällskapet

Skogsgläntan 16S-435 38 Mölnlycke

SwitzerlandDeutsche Cichliden Gesellschaft

Am Balsberg 1CH-8302 Kloten

CICHLID ORGANIZATIONS WORLDWIDE

Taiwan (R.O.C.)Taiwanese Cichlid AssociationNº17,Lane 239,An-Ho Road

Taipei

United KingdomBritish Cichlid Association

100 Keighley RoadSkipton, North Yorkshire, BD23 2RA

U. S. A.American Cichlid Association

P.O. Box 32130Raleigh, NC 27622

Adv. Cichl. Aquarists South CaliforniaP.O. Box 8173

San Marino, CA 91108

Apistogramma Study Group1845 Jaynes Road

Mosinee, WI 64455

Cichlasoma Study Group6432 South Holland Court

Littlerton, CO 80123

Fort Wayne Cichlid Association9638 Manor Woods Rdf.

Ft. Wayne, IN 46804

Greater Chicago Cichlid Association2633 N. Rhodes River

Grove, IL 60171

Greater Cincinnati Cichlid Association15 W. Southern AvenueCovington, KY 41015

Illinois Cichlids and Scavengers7807 Sunset Drive

Elmwood Park, IL 60635

Michigan Cichlid AssociationP.O. Box 59

New Baltimore, MI 48047

Ohio Cichlid Association3896 Boston Rd.

Brunswick, OH 44212

Pacific Coast Cichlid AssociationP.O. Box 28145

San Jose, CA 95128

Rocky Mountain Cichlid Association5065 W. Hinsdale Cir.Littleton, CO 80123

Southern California Cichlid AssociationP.O. Box 574

Midway City, CA 92655

Texas Cichlid Association6845 WinchesterDallas, TX 75231

4

99ISBN 3-928457-05-5

Xenotilapia spiloptera, photographed by Ad Konings at Mabilibili, Tanzania.

Retroculus lapidifer from the Rio Tocantins. Photo by Ron Bernhard.

The CICHLIDS yearbook