Cichlid Room Companion


The last minutes of speciation

By , 1991. printer
Martin Geerts, 2011

Classification: Evolution.

Cynotilapia sp. 'Chinyankwazi' Cynotilapia sp. "Chinyankwazi" in reef between Chinyamwezi Island and Nankhumba peninsula. Photo by Ad Konings.

The family Cichlidae is exceptionally rich in species. Of course, this fact is caused by the ease with which these fishes are able to speciate. Moreover, cichlids are also known for their plasticity, which means that within one species considerable phenotypic variation can occur. It is more than justified when the South African ichthyologist Michael Bruton (1989) states: 'The adoption of alternative phenotypic states in nature is probably more widespread than is currently realised and many populations which are currently recognised as species may in fact, be no more than ecophenotypes of one or another homeorhetic state". It might thus be clear that the question whether one or another population or group of populations should be regarded as a distinct species or not, will be a hotly debated issue among cichlid-scientists for many years to come.

Evolutionary biologists usually call attention to the cichlids of Lake Nabugabo whenever they want to demonstrate that speciation in these fishes can take place in a relatively short period of time. Lake Nabugabo was part of Lake Victoria in previous times. About 4000 years ago a bay was separated from the main lake by a sandbar. In the lake thus formed Greenwood (1965) recognizes five distinct species which must have developed from ancestral Victorian populations which remained in the isolated bay. Since Greenwood's paper the cichlids of Lake Nabugabo have been presented in many publications as examples of how much faster speciation might progress than was previously accepted.

Some recent observations, however, suggest that speciation among cichlids may operate at an even faster speed than the cichlids of Lake Nabugabo let us assume. One of these observation relates to the description of a new Mexican cichlid (Werner & Stawikowski, 1988). The new species, which is named Paratheraps breidohri by the authors, is reported to be endemic to Presa de la Angostura, a reservoir in the southern part of México. The closest relative of P. breidohri, named Paratheraps hartwegi by the same authors, occurs in the rivers and streams flowing into the reservoir as well as in the reservoir itself. Hence, we cannot escape the notion that P. breidohri developed from a population of P. hartwegi after the man-made dam formed the reservoir. This would mean that P. breidohri, if accepted as a distinct species, would just be less than 30 years old! On the other hand we cannot exclude the possibility that P. breidohri should be regarded as the lacustrine morph of P. hartwegi. Further examination is required to provide a solution to these questions.

Very recently it became apparent that human intervention may have yielded a new species in Africa as well. This species was named Haplochromis erythromaculatus by the authors, De Vos et al. (1990). H. erythromaculatus is found in Lake Bulera and Lake Ruhondo. These lakes lie in the northern part of Rwanda where, as early as 1907, the German H. Schubotz investigated the fish fauna of Lake Ruhondo. He concluded that this lake was inhabited only by small barbs, but certainly did not mention cichlids. Later fishery biologists tried to enrich the fish fauna of the Rwandan lakes with species suitable for consumption. Therefore they introduced several tilapiines which were collected in Lake Edward. Among the tilapias there were apparently also some specimens of a haplochromine. From these "contaminating" individuals H. erythmmaculatus must have been originated (in a period of time which is less than 80 years). De Vos et aL tried to discover what species in Lake Edward was transferred to Lake Bulera and Lake Ruhondo, but could not find it. The cichlids of Lake Edward are apparently not well studied yet. Nevertheless, the recognition of H. erythromaculatus as a valid species suggests that cichlids are able to speciate much faster than previously thought by evolution biologists. Jos Snoeks (pers. comm.), one of the authors, believes that speciation needs more time than the suggested 80 years.

Not only do human interventions in nature allow us to study the speed of speciation among cichlids, but also our expanding knowledge about the ecological history of the waters in which cichlids live contributes to these studies. Moreover such knowledge suggests that speciation among cichlid can take place at a much higher speed than is generally accepted.

Recently, Owen et al. (1990) showed that the level in Lake Malawi fluctuated more frequently and between much higher extremes (not the yearly fluctuations) than had been concluded from previous studies. From the paper by Owen et al. it appears that the level of Lake Malawi, in the last decades of the eighteenth century and in the first decades of the nineteenth century, was at least 120 meters below that of today's lake. If these findings prove to be true it would mean that about 150 to 250 years ago many of the locations which are presently inhabited by endemic Mbuna were a part of the continent. This is true for the islands Likoma, Chizumulu, Mbenji, the Maleris, Boadzulu, Thumbi West, etcetera. The authors conclude that the Mbuna, which are endemic to these islands, cannot be older than the mentioned 150 to 250 years (see Ad Konings' note). For the time being we regard these endemic species as valid species and not as ecophenotypes.

Now, examining the bathymetric map of Lake Malawi, we should be able to indicate from which locations the endemic species of certain islands were derived and which species can be regarded as their ancestors.

In the frame of this writing it is, unfortunately, not feasible to continue the discussion of these examples or to detail others. Cichlidists with a multidisciplinary background will, without doubt, be able to find more and maybe better examples. Then it will be possible to gain a better understanding of the process which is called speciation, of which it is usually thought that a man's lifetime is too short to see it happen.


  • Bruton, M. (1989) The ecological significance of alternative life-history styles. pp. 503-553 in M. Bruton (ed.) Alternative life-history styles of animals. Dordrecht, Netherlands.
  • De Vos, L., Snoeks, J. & Thys V. D. Audenaerde, D.F.E. (1990) Description d'Haplochromis erythromaculatus, espece nouvelle (Teleostei, Cichlidae) des Lacs Bulera et Ruhondo, Rwanda. Ichthyological Eplorations of Freshwaters Vol. 1 (3); pp 257-268.
  • Greenwood, P.H. (1965) The cichlid fishes of Lake Nabugabo, Uganda. Bull. Br. Mus. nat. hist. (zool), 12; pp 315-357.
  • Owen, Rb., Crossley, R., Johnson, T.C., Tweddle, D., Kornfield, I., Davison, S., Eccles, D.H. & Engstrom, Dz., (1990) Major low levels of Lake Malawi and their implications for speciation rates in cichlid fishes. Roc. R. Soc. London. B-240; pp 519-553.
  • Werner, U & Stawikowski, R. (1988) Ein neuer Buntbarsch aus Südmexico: Paratheraps breidohri gen. nov. spec. nov. DATZ 41 (1); pp 20-23.
(This article was originally published in "The Cichlids Yearbook 1" Cichlid Press; pp. 94-95. It is here reproduced with the permission of author Martin Geerts).

Ad Konings' note

In the paper of Owen et al. (1990) it is concluded that most of the endemic species inhabiting islands, which were part of the mainland 200 years ago, have developed in recent times. Disregarding the fact that the definition of a species is still hotly debated and will probably never become an universally agreed entity, there are some species (distinct taxa) which are found on both sides of the once dried-up southern part of Lake Malawi. This would suggest that, during the low lake level, there must have been a suitable habitat where at least these species could have survived the "drought". These species are P. barlowi (distribution: Mbenji, Maleris, Thumbi West, Nkhudzi and Eccles Reef). P. sp. 'Dumpy" (distribution: Maleris and Fort Maguire), P. sp. 'Zebra Red Dorsal" (distribution: Mpanga Rocks, Nakantenga, Nkudzi and Eccles Reef) and Copadichmmis azureus (distribution: Nkhomo Reef, Mbenji, Maleris and Eccles Reef).

One of the habitats that could have served as a "sanctuary" during a low lake level is a reef between Chinyamwezi Island and the Nankhumba peninsula. This reef remains at least 24 meters below the surface while its basis meets the sand at a depth of at least 105 m. I have visited this reef in December 1990 and found Cynotilapia sp. "Chinyankwazi" in vaste numbers (see photo). Also present are P. tropheus, which is further distributed around the Nankhumba peninsula and Chinyamwezi, P. flavus, which is known from Chinyankwazi and P. ater, known from Chinyankwazi and Chinyamwezi.


Geerts, Martin. (April 18, 1998). "The last minutes of speciation". Cichlid Room Companion. Retrieved on December 16, 2018, from: