The Tree of Earthworms

Earthworm taxonomists describing what they do to a layperson is hilarious to watch. Laypeople often have a difficult time understanding the concept of a species – you will regularly hear statements that there are only 50 insect species, for example.

Insect species often differ in colour and patterning, so it’s easy to then correct a layman’s misconceptions about insect taxonomy. Now imagine trying to do the same for earthworms, all pink worms wriggling in your garden soil with no really conspicuous variation. In the eyes of most, there is just one earthworm species in the world.

Of course, that’s ludicrous. Earthworms (family Lumbricidae) comprise around 300 species, all characterised by the presence of a multilayered clitellum, a darker, thicker ring of segments near the head that plays a role in reproduction.

Defining species requires close looks at the prostomium (very first segment), the patterns of setae on the body, the clitellum’s size and position and its tubercula pubertatis (a swelling on the bottom of the clitellum), and the spermathecae. But it is true that they are still very similar to each other, and so the taxonomy and phylogeny of earthworms is a big black box, despite earthworms’s supreme importance in agriculture and ecology (90% of a temperate soil’s biomass is earthworms!). For example, the number of proposed genera ranges from 6 to 45.

This amount of confusion is why I think it’s important to highlight this new paper:

ResearchBlogging.orgDomínguez, J., Aira, M., Breinholt, J., Stojanovic, M., James, S., & Pérez-Losada, M. (2015). Underground evolution: New roots for the old tree of lumbricid earthworms Molecular Phylogenetics and Evolution, 83, 7-19 DOI: 10.1016/j.ympev.2014.10.024

This is, to my knowledge, the most extensive and robust phylogeny for the Lumbricidae done to date, analysing the relationships between 86 species from 28 genera. It does both a molecular phylogeny and a morphological one, although external morphological characters proved to be too homoplastic and unreliable, so only the molecular phylogeny is considered here.


Besides the phylogeny, other insights acquired from this new phylogeny are:

  • The ancestral lumbricid was hermaphroditic;
  • The ancestral lumbricid probably had two spermathecal pairs (Rcent ones have anything from 0 to 8);
  • The ancestral lumbricid was endogeic (soil-dwelling), with multiple independent evolutions to epigeism (leaf-litter dwelling) and anecism (making permanent vertical burrows in the soil);
  • We really, really need to work harder on lumbricid taxonomy;
  • Earthworms are crap at dispersing – genealogy and geography are tightly correlated in a lot of the clades.

Finally, they also did a molecular clock analysis, which found origination in the lower Cretaceous (125Ma) of Europe and a Miocene period of diversification. However, given that no fossil calibrations were used (no fossil lumbricids are known, so this isn’t the researchers’ fault!), these estimates should be taken with a  grain of salt (see my summary of the molecular clock for more info).

Read the paper for all the details (or ask me for a deeper discussion). This is the current state of the art for lumbricid systematics and phylogeny, and future research will need to involve more species, and more morphological characters. In my opinion, internal anatomy and glandular organs should be concentrated on, any features that are intensively used by earthworms and that vary between different earthworm ecotypes (e.g. structures and chemicals involved in reproduction, or the digestive tract). Of course, these are much trickier to investigate, requiring dissection, histology, and chemistry sets.

Leave a Reply