What determines speciation at a molecular level?

At what point does a scientist determine two lineages are different enough to be considered separate species?

Does it have a margin of error?

  • $\begingroup$ FYI, the term genetic speciation does not exist. It is just speciation. But one can use genetic tools to estimate divergence between lineages and eventually base its concept of species on those data. $\endgroup$
    – Remi.b
    Aug 24, 2015 at 15:16

3 Answers 3


As you probably know, there exists many different species concepts (or definitions of species). Often, you see a separation into at least six different categories of species concepts, which are then often subcategorized futher (and they can also overlap to some extent), namely:

  • Biological species concept
  • Phylogenetic/cladistic species concept
  • Evolutionary species concept
  • Ecological species concept
  • Cohesion species concept
  • Morphological/Phenetic species concept

Broadly, these groupings show up in different forms in e.g. Futuyma (1998), DeQueiroz (2007) and Mallet (2007). The Wikipedia page on species concepts also provides a useful overview.

The biological species concept is probably the most well-known, but is very difficult to apply for many organisms, due to for example lack of knowledge (we know very little about the reproductive biology for most species of earth), their breeding system or maybe because we are dealing with extinct organisms. Your question is most closely connected to the phylogenetic/cladistic group of species concepts, which aims to delimit species based on lineages/clades/monophyletic groups, often using molecular phylogenetics for inferences.

I don't know the literature of the phylogenetic/cladistic species concept well enought to answer your question fully, but I know that there exists multiple ways to delimit species within this group of species concepts. The most "extreme" forms/interpretations state that a species can be any monophyletic group with inhereted characteristics that you can diagnose, and this can lead to very narrow species definitions and "splitting" of one species into several. However, more conservative definitions look at e.g. monophyletic gene trees, where several geographic subspecies can be grouped withing the same species.

All-in-all, I don't think it is possible to give a definite answer to your question "At what point does a scientist determine two lineages are different enough to be considered separate species?". It depends on the species concept that is used, and how this is interpreted by the taxonomists conducting a specific study. Also note that phylogenetic/cladistic species concepts are related to the idea of Least-inclusive taxonomic units (LITUs, see Pleijel & Rouse, 2000), that aim to partially separate "species" from taxonomy and instead only consider taxonomic units, based on current knowledge.


  • 1
    $\begingroup$ @AliceD I don't know about superior, but maybe a bit more comprehensive. Hopefully, somebody that is actually involved in taxonomic work using molecular methods can provide a more specific and detailed answer. $\endgroup$ Jun 1, 2015 at 14:06

Speciation in sexually-reproducing organisms can be identified as the inability to produce viable offspring. In other words, when two suspected sub-species are not able to produce viable offspring, they can be considered to be two species. In this case it is not so much determined at the molecular level, but at the organismal level.

This mating procedure does not work in asexually reproducing species. In this case molecular techniques are needed. Navarro & Barton (2003) investigated what the molecular differences were induced by evolution by comparing the genomes of humans and chimpansees. These two species diverged relatively late in evolution and their DNA differs by about 2.1%. It turned out that the main differences between the two genomes were clustered around regions where chromosomes were rearranged. Although indicative of the cause of speciation, it can nonetheless also be the result of it. We can never be sure.

- Navarro & Barton, Science (2003); 300: 321-4

  • 1
    $\begingroup$ AFAIK, the test for speciation is a genetic cross, or mating. If two individuals can mate to produce fertile offspring then by definition they are both members of the same species. The converse is also true (at the population level). Back in the 1970's experiments were done where fruit fly siblings were separated (reproductive isolation) for 10 generations, the resulting progeny (10th cousins), could no longer cross-breed. $\endgroup$
    – mdperry
    May 27, 2015 at 22:50
  • 2
    $\begingroup$ I think this answer is taking a much too narrow view. Most taxonomic work is not using the biological species concept in operational terms (but it might be assumed implicitely), instead focusing on molecular phylogenies and the morphology of organisms. It is seldom that the biological species concept can actually be tested, even though it is important from an evolutionary perspective, and for most named/described species on earth the ability to interbreed has not been explicitely considered or tested. $\endgroup$ Jun 1, 2015 at 13:34

The genus Rhododendron has been divided into some 900 plus species. Often these species hybridize in nature creating what are called hybrid swarms. Although most Rhododendron species are diploids there exists tetraploid, hexaploid, octaploid, and decaploid Rhododendron species. Contact zones between Rhododendron species containing diploid and tetraploids often result in triploids. Some of these triploids are semifertile and able to interact with both the diploid and tetraploid populations.

Man made hybrids between Rhododendron species can be very complex and multi-generational. Man made hybrids have created triploids and tetraploids using only diploid species. Hybrids can be created between the origin diploid species and the man made triploids and tetraploids.

Genetic studies of closely related Rhododendron species suggest that at least some evolution in Rhododendron is reticulated. In other words the barriers for sexual reproduction between species are often not complete. Moreover these lack of absolute sexual barriers actually allow gene flow between species even when species are of differing ploidy levels.

The concept of species implies separation. Yet often when one examines closely related species (populations) in genus such as Rhododendron, the separation sexually is not distinct and absolute. The terms species and hybrids blend when one introduces the concept of reticulated evolution in that the hybrid swarms become a potential source for new species.


You must log in to answer this question.