It may seem like a trivial question but it is far from being so. Even a few years ago scientists diverged widely in their estimates on the number of cichlid species. Numbers ranged from 1,800 to over 3,000. In his wonderful book “The Cichlid Fishes: Nature’s grand experiment in evolution” (2002), George W. Barlow (1929 – 2007), a noted American Scientist and a brilliant mind who I had the pleasure to meet and more intimidatingly, have in the audience of one of my cichlid talks (thankfully approvingly nodding, smiling and commenting all along, so I felt reassured), gave a few quotes that should illustrate the problem. He quoted a scientific paper in the introduction of his book “One large lake of Africa alone may have from 500 to 2,000 species of cichlids, though nobody knows the real count”, how about that for a range?
To understand the problem we first have to take into consideration what is a species. From that starting point we face our first big major obstacle, as there are several different species concepts (e.g. biological species concept, evolutionary species concept, cladistic species concept, phylogenetic species concept, and more) and even considering just once species concept, let’s say the biological species concept, which is probably the most widely accepted, we often disagree on what a species is. Using the BSC Ernst Mayr defined a species as "species are groups of interbreeding natural populations that are reproductively isolated from other such groups."
If we just consider geographical isolation, what happens when one cichlid species inhabits two river systems, a barrier which prevents the two populations from interbreeding naturally? Are these populations then different species? It seems that some modern taxonomists tend to think so. If we, however, consider reproductive isolation as the differentiating mechanism between two populations that generally prevent them from interbreed, then we can fit cichlid populations living in two or more isolated bodies of water as a single especies. We however face another problem because we now know that in crosses between many different cichlid species fertile offspring can be produced, even in crosses between individuals considered in different genera!
To complicate things further recent discoveries have shown that a lot of the diversity of species is due to what is known as reticulate evolution, where one lineage (e.g. species) originates through the partial merging (hybridization) of two ancestor lineages. One case of reticulate evolution is known as “hybrid speciation”, which is a form of speciation where hybridization between two different species leads to a new species, reproductively isolated from the parent species. A species in mind that exemplifies this case has been found to be Amphilophus istlanus, a popular Mexican cichlid, which is likely to be the product of hybridization between two lineages represented today by Mayaheros beani and Amphilophus trimaculatus (Říčan et al., 2016). As we can see, speciation does not follow a pattern that we can accommodate in our limited and easy to order and visualize classification schemes.
With this in mind, let’s come back to the question of how many cichlid species are there. Let’s first consider legally valid species. The International Code of Zoological Nomenclature (ICZN), put together by the International Commission on Zoological Nomenclature, is a widely accepted convention in zoology that rules the formal scientific naming of organisms treated as animals. So for an animal species to be valid, it has to be named in rigorous compliance with the latest version of the ICZM, which is version 4, in force since 1 January 2000 nowadays.
The fact that I have created and for the last 22 years edited “The Cichlid Room Companion", which lists every single cichlid species ever described, gave me the opportunity to make a good compilation of this otherwise very difficult to tally group of papers. It happens that there are 2,249 published cichlid descriptions to this day that comply with the ICZN. However, 12 of those descriptions correspond to species which cannot possibly be identified, this happens because there is not a “voucher” for the species, meaning a drawing, preserved specimen, or picture of the representative of that species and the description of it is too vague to be identified with any living organism. So excluding those we have to keep the number at 2,237 species.
But wait, as it happens some of those published descriptions are considered — either by the original author(s) or by subsequent reviewers — as sub-specific taxa. That means that the differences between some described species are so small as not to be considered different but rather geographical variations of one single species. We have to be cautious here as although the publication of subspecies is regulated by the ICZN, this is a subjective decision based on the opinion of the person doing the publication or revision. An example of this case is the ten subspecies of Mayaheros urophthalmus described by Carl Hubbs in 1936-1938. Subsequent workers may decide that some subspecies should be considered merely synonyms of the nominal species or that they should be elevated to a specific status of their own. But again this is a subjective decision based on the opinion of the reviewer, and different opinions may conflict with each other. Currently there are 47 cichlid taxa generally accepted as subspecies (not counting their nominal species), which if subtracted from the 2,237 species, left us with 2,190 formally described cichlid species.
However, not all of those 2,190 species should be considered as valid. There are many cases where two descriptions concern to the same species, in which case they are considered synonyms, with just one of them valid. There are basically two types of synonymies: objective and subjective.
Objective synonymies are straightforward but small in number, since they refer to the specific case in which the same specimen (type) was used to describe two different species. Obviously, just one remains and the second one is an objective synonym. Using the ICZM principle of priority, the one that remains is that published first.
Subjective synonymies are much more conflicting, since they are based on the opinion of one author — or group of authors — that consider that a described taxon (species) is the same as another previously described taxon, since in their opinion the differences between them do not justify their recognition as different species. The ICZN does not regulate such opinions, which can be published in any form, and followed (or not followed) by different persons.
Different authorities (e.g. Fishbase or the Catalogue of Fishes of the California Academy of Sciences) have different criteria. In the Cichlid Room Companion catalogue, which is sanctioned by our own set of specialists, we lists 464 validly described taxa in synonymy. If we subtract these from the 2,190 species for which we carry the sum, we are left with 1,726 valid species.
If we consider that according to FishBase, 33,100 species of fish had been described by April 2015, with an average of 5 species per family, we can infer the evolutive success of the Cichlid family, which alone accounts for more than five percent of all fishes!
There are also species that are recognized by some authors as “potentially undescribed species” which means that they have reasons to believe the form in question should be recognized at one point as a new species and get a formal (ICZN compliant) description. As expected, the number of such potentially recognizable taxa is extremely variable, and ranges from forms for many years well known and with a general agreement of their uniqueness, to forms which cannot even be told in what they are different from others by their same proponents. This number changes every day. In the Cichlid Room Companion we have as criterion to recognize such potentially undescribed species just when they are given a temporary diagnosis (traits that make the form different from existing valid species) to its closest relative, and a publication of the form temporary name (often between quotation marks) exists. With this criterion we list 656 potentially undescribed species, which would raise the number of cichlid species to 2,382. With this number of undescribed species however, I am very cautious to offer a range of error. With the many different forms Lake Malawi and Lake Victoria have, they are the champions in the tally for such potential species.
For some of the potential species the decision of whether to describe them or not is not simple and requires a previous knowledge of their distribution and variability, possibly with a comparison of their DNA using new techniques, before a decision can be reached. Take for example the so-called Honduras red-point cichlid, a variant of the convict cichlid from Honduras. This form was discovered by Rusty Wessel, and there has been quite some discussion about whether this species has to be considered, or not, a geographical variant of Amatitlania nigrofasciata (the convict cichlid) or a valid species. New field evidence suggests that this species occurs together with the normal A. nigrofasciata and in fact form separate breeding pairs, but some DNA information contradicts this view. More information is hence needed.
In recent times we have seen an emerging situation in regard to the publication of new species descriptions. In the past, synonymies were mostly published by mistake. Because of slow communication, limited information on the specimens used, limited access to museum collections or other problems, two reputable authors may have described the same species twice as a result of these problems. Nowadays, however, with the overwhelming majority (though likely not all) of the different cichlid forms already known, there is little room left for discovering new previously unknown species which are clearly unique.
New synonymies emerge from the sometimes eager desire of authors to publish descriptions of new species based on the slightest differences between them. Such descriptions often do nothing more than confuse the general public, since people can often not discriminate between the different species described, and sometimes not even the authors can without detailed analysis or knowledge of their provenance. This is an unfortunate situation since the recognition of the natural variability of a species is ignored. In some occasions, the authors of such descriptions exhibit a profound lack of knowledge of the basic variability of the species in question which they deem as new.
With this text, I hope I have given you a clear overview of some problems facing species recognition as well as an estimated number of valid cichlid species available. It is clear that taxonomic work is never to be finished, but the good news is that we will be learning more and more about our beloved cichlids every day. And after all, species classification is a human construct intended to facilitate the study and understanding of biological variability, nature is not concerned with it at all.
- Barlow, George W.. 2002. "The Cichlid Fishes (Nature's Grand Experiment in Evolution)". Perseus Publishing. pp. 352 pp. (crc03927) (摘要)
- Říčan, Oldřich & L. Piálek, K. Dragová & J. Novák. 2016. "Diversity and evolution of the Middle American cichlid fishes (Teleostei: Cichlidae) with revised classification". Vertebrate Zoology. v. 66(n. 1), pp. 1 – 102 (crc07292) (摘要)
© Copyright 2019 Juan Miguel Artigas Azas, all rights reserved
Artigas Azas, Juan Miguel. (一月 10, 2019). "How many cichlid species are there?". Cichlid Room Companion. 检索 七月 19, 2019, 从: https://www.cichlidae.com/section.php?id=299&lang=cn.