Cichlid Room Companion

Wonderful cichlids

Establishing relationships

Par , 2019.
Dernière mise à jour le 31-oct.-2019

Juan Miguel Artigas Azas, 2010

" A summarized view of the different techniques used to obtain the phylogenetic trees depicting how evolution happened "

It has been a long working project for taxonomists to reconstruct the tree of life, in which all species are shown as deriving from an ancient common ancestor in an evolutionary pathway where specializations necessary to survive, and more importantly thrive via natural and sexual selection, have produced the rich mosaic of life we see all around us. Most of the species that have existed have been left behind and just the few more evolved to thrive in the current environmental conditions are still present.

Up until the late 1990s all phylogenetic relationships among species were established by evaluating the presence of some specialized characters shared between some of them; those characters were inferred to have been inherited from a common ancestor and hence the species possessing them were considered to be related. The more specialized characters were common to two or more species, the more closely related those species were supposed to be, since they had like;y separated from a common ancestor more recently than those sharing fewer of those specialized characters.

Phylogenetic systematics

The formal name given to the field within biology that intends to reconstruct the evolutionary history of organisms based on the characters they share with their last common ancestor is Phylogenetic Systematics, also known as Cladistics. What Phylogenetic Systematics intends is to recreate the species phylogeny, which is the history of the evolution of a group of taxa (such as species) from their common ancestors. This includes the order from which they branched off from each other and, when possible, when that happened.

Mâle dans son habitat Neetroplus nematopus mâle dans le centre du Costa Rica. Photo de Juan Miguel Artigas Azas. identifié par Juan Miguel Artigas Azas
Individu en aquarium Neetroplus nematopus (above) and Hypsophrys nicaraguensis (below) are sister species, but their uniqueness granted them a each monotypic genus. Photo de Brian Traas. identifié par Juan Miguel Artigas Azas

The representation of the evolutionary pathways is made with the use of what is known as evolutionary or phylogenetic trees. Evolutionary trees are diagrams showing the evolutionary relationships between different animals: their phylogeny.

The different branches of a phylogenetic tree are created with groups of taxa that share a common derived (i.e. evolutionary new) feature not shared by other closely related groups, referred to as an apomorphy. An apomorphy is necessary to diagnose a new genus or any other taxonomic group, in which all of its members must have this apomorphy, or shared trait. For example: The anal fin in male Pseudocrenilabrus shows a reddish posterior tip rather than the round, often ocellated, eggspots seen in related genera.

Unlike an apomorphy, a plesiomorphy is an evolutionary ancestral trait that is present in a group of individuals, but it is not unique to members of that group. For example: all cichlids have fins, but we cannot diagnose cichlids as having fins because most other fish also have fins, a plesiomorphic character. When a phylogenetic group is defined, being for example a family or a genus, it is generally accepted that they have to be monophyletic. A monophyletic group is made of related taxa that have been produced by one common ancestor. Monophyletic groups consist of species more closely related to each other than of any of them with a species outside the group.

In monophyly we cannot have (for example) nested genera, which are known as paraphyletic. A group is paraphyletic if it consists of the group's last common ancestor and all descendants of that ancestor excluding a few monophyletic subgroups. I give you an example: Říčan et al. in 2016 validated the previously described Central American cichlid genus Amatitlania (altoflava, kanna, myrnae, nanolutea, nigrofasciata, sajica, semptemfasciata and siquia), which was considered a junior synonym of Cryptoheros (chetumalensis, cutteri and spilurus) because they found that the monotypic genera Hypsophrys and Neetroplus were nested in it, turning them paraphyletic. Species in the genus Cryptoheros evolved from one common ancestor with that of all of the species in the genera Amatitlania, Hypsophrys and Neetroplus. When this was discovered, Říčan et al. had two choices: either to lump the latter three genera in Cryptoheros (making it monophyletic) or to divide Cryptoheros in the four genera to recognize the four monophyletic lineages. Since they considered that the distinguishing characters of Hypsophrys and Neetroplus were important enough to be recognized as new evolutionary lineages, they decided for the latter.

Problems in establishing relationships

Unlike paraphyly, a misleading effect when constructing a phylogenetic tree is that of polyphyly. Polyphyly is a term used when a group of similar taxa have actually been produced by more than one common ancestor. This can happen for example when a given phylogeny is not yet fully understood and a particular genus is used as a “catch-all” recipient for a yet poorly understood similar group of species. This has happened a number of times in cichlid taxonomy, for example in genera like Lake Malawi Pseudotropheus, currently with 46 species (28 potentially undescribed). The genus has been used for Lake Malawi cichlids with a slender body and a rather stretched and convex snout, together with a pattern of vertical bars. The genus it is currently viewed as polyphyletic, with probably just a small group of species related to the type species of the genus: P. williamsi, being monophyletic (e.g. Tawil, 2011). More genera will have to be defined to accommodate for those polyphyletic groups within Pseudotropheus.

When dealing with classical phylogenetic systematics, one of the problems early ichthyologists were faced with was that some evolutionary pathways were poorly understood. One of the major forces, that of convergent evolution, was not known at the time. Convergent evolution is the independent development of similar features in species of different lineages (not closely related to each other) — i.e. similar problems can lead to similar solutions — which could be misleading in cladistics. The term for a feature that two species share for any reason other than common ancestry is called homoplasy. For example, birds, bats, and insects all use similar wings to fly although they are not closely related, those were developed independently by random mutations converging in the same solution, and in these cases the homoplasy is the possession of wings.

An example closer to home refers to the initial placement by Jordan et al. (1899) of Herichthys carpintis in the genus Neetroplus due to their similar dentition. Both species: H. carpintis and Neetroplus nematopus, although not closely related, have spatulate or chisel like teeth useful for cutting vegetation, which at the time was thought to be an apomorphic character. Herichthys carpintis however belongs to the phylogenetic group of the so-called Herichthyines and N. nematopus to the Amphilophines. Both groups are the result of separate evolutive lineages that diverged a long time ago. Both species have independently evolved a similar shape of teeth for a similar use. Another example are the three-tipped teeth of Petrochromis from Lake Tanganyika and Petrotilapia from Lake Malawi. Those teeth really are identical but those genera are not at all related to each other.

Convergent evolution is present everywhere, not just in morphology but also in behavior. Think about mouth-brooding cichlids, which developed this fry-protecting behavior independently in America (e.g. Heros liberifer, Apistogramma barlowi, and others) and in Africa (all Malawi and Victoria cichlids). When convergent evolution was not taken into consideration in classical phylogenetic systematics, many puzzling situations could not be explained and errors were made in phylogenetic trees.

Also causing problems in phylogenetic systematics is adaptive radiation. Adaptive radiation is the opposite of convergent evolution, and according to Wikipedia is “a process in which organisms diversify rapidly from an ancestral species into a multitude of new forms, particularly when a change in the environment makes new resources available, creates new challenges, or opens new environmental niches”. Adaptive radiation can similarly cause many problems when evaluating different features of closely related species. Two apparently very distinct forms may be in fact very closely related. Take for example the already mentioned Neetroplus nematopus and Hypsophrys nicaraguensis in Central America, they are sister species (derived from a common ancestor not shared by a third species) and still they are morphologically so different that they are currently placed in different genera!

Another effect, only recognized in recent years, is that hybridization, or reticulate evolution, can lead to the development of intermediate forms (hybrids) of unrelated species that acquire a particular adaptation that allows them to thrive and become, what we now call, a new species. Such species may also create taxonomic problems since they share features of both parental species and their relationship to either lineage can be uncertain. This is particularly troublesome in molecular phylogenies because formation of intermediate hybrid taxa is rare. A common effect of hybridization are conflicting results of different data sets, e.g. between mtDNA and nDNA are between DNA and morphology.

One recent example of such a situation concerns a rheophilic morph of Vieja hartwegi from the upper Rio Grijalva (Gómez-González et al., 2018). The apparent sudden appearance of this form was puzzling since this area had been widely researched and this particular form had not been seen before. Moreover, the new form significantly differs in several aspects of its morphology from any other reophilic cichlid found in Central America, and, worse yet, it does not comply with the diagnosis of Vieja! A detailed study (in preparation) by Oldřich Říčan and Rico Morgenstern suggests with strong argumentation and circumstantial evidence that this form may in fact be just a natural hybrid between Vieja hartwegi and Chiapaheros grammodes. Let’s hope nobody jumps the gun and decides to describe a “new” species before the whole situation is carefully studied, as this form would be difficult to classify in a genus. Worse yet, if we consider the current trend in Central American cichlid taxonomy, it could even get its own genus!

In spite of all this, early ichthyologists presented us with a very close general picture of group relationships, and described for us most of the species we now know. We should really be amazed and thankful for the detailed and fantastic work early taxonomists did and the knowledge they provided us; as Isaac Newton once said: “if I have seen further it is by standing on the shoulders of giants”. We should also be thankful to the European countries that supported the initial exploration and research necessary to obtain the thousands of specimens needed to start putting together the puzzle of the natural world all around the globe.

Mâle à Chizumulu Pseudotropheus williamsi mâle de l'île de Chizumulu, lac Malawi [Malawi]. Photo de Ad Konings. identifié par Ad Konings

The DNA based phylogeny

Beginning in the early 1990s, papers started to be published that made use of mathematical methods comparing not apomorphic features, but segments of the species DNA. The statistical methods of multivariate analysis and the power of computers made this possible. The idea is to compare each pair of selected segments of DNA in different organisms — the technology locates, isolates, and reads these segments — and then reconstruct possible changes that could have happened over time through random mutations in these DNA segments, creating phylogenetic trees. It is known approximately how many mutations can be expected over a certain time period, so by counting the differences (mutations) in the DNA, and comparing that with the same piece of DNA of a far-away relative (a control species), the age difference of the species examined could be estimated.

Initially it was common to use segments of mitochondrial DNA (mtDNA) for these studies but now we have learned that in many occasions that may be of little use, since mitochondrial DNA is inherited from the mother alone and it is prone to misleading results when hybridization has happened in the past — a phenomenon that, we are currently learning, has played an important role in evolution. The ideal situation for obtaining a solid phylogeny is to select for comparison DNA segments that have not recently been influenced by evolutionary pressures, but those that have naturally drifted with time. This would allow a better estimation of the time that has lapsed since the separation from a common ancestor. Since about 98% of vertebrates DNA is non-coding (not reflected in morphology), the base for selection is wide!

In recent years the so called Next-Generation ddRAD (double digest Restriction Associated DNA sequencing) analysis that uses most of the nuclear genome (nDNA) for comparison, has become an affordable technique due in part to the power of modern computers. This technique has allowed for acquiring, comparing, and processing an astounding number of DNA base pairs (DNA building blocks). Modern cichlid phylogenetic studies use about 750 million base pairs (e.g. Říčan et al., 2016).

The ddRAD technique is not without problems, but DNA analyzing techniques are advancing. Both the acquisition and selection of the DNA sequence to compare, and the power and accuracy of the mathematical algorithms used, are improving every day, offering us a better picture of evolution. We should, however, never forget that we were not there when speciation happened; we could not observe the natural forces that drove it and we could not see the species that became extinct along the way, the hybridization, isolation, dispersion, and other events that eventually produced the wonderful fauna that we can still observe and try to understand today.

Références (4):


Artigas Azas, Juan Miguel. (octobre 31, 2019). "Establishing relationships". Cichlid Room Companion. Consulté le décembre 13, 2019, de: