How species are identified


The naming of a new species is often taken to be a significant event in biology, much excitement in the media is devoted to the identification of a new species, but the truth is that it is mundane. Certainly some things are more rare than others (new mammals tick along at a rate far, far below that of new wasps for example) but the event itself is pretty commonplace. By the best estimates, biologists have identified something like two million unique species, which is quite a few by anyone's measure, but the total number is quite probably ten times that or more. Quite simply, it will probably take us another century or three to identify every species currently alive (though of course the rate of extinction is such that there are plenty we won't have to deal with as they will have died out before we even find out they ever existed).

Taxonomy, the field of identifying species, has been sniffily dismissed as mere stamp collecting, but this attitude belies an ignorance of what it means to correctly identify species. These are the fundamental units of biology, much as elements are for chemistry. Imagine the complexity of trying to work out chemical reactions and likely patterns and processes without recognising that some substances were composed of a single element, and others were mixtures or compounds. If we cannot identify and separate out species, the rest of biology is rather left floundering.

This might still seem like something only of interest and relevance to academics, but the implications are much more broad. A farmer needs to know which weed it is that has invaded his plot and what pesticide to use, or which flies might blight his livestock. If you're ever bitten by a snake or spider you should hope the species can be correctly identified or the anti-venom administered may be incorrect. That plant may yield a new anti-bacterial drug, but are we sampling one species or two that look alike? A new mosquito is spreading, but is it a malaria carrying species or not?

Written By: Dr Dave Hone
continue to source article at


    • In reply to #1 by Stafford Gordon:

      This is the sort of thing I come here for mainly.

      Although now that I’ve learnt a bit about how science works and this article is within my grasp I still enjoyed reading it.

      Yes, I wish there was more general discussion of science.

      I would encourage anyone who is interested in developing taxonomies (of species or anything else) to check out the Protege tool from Stanford:

      Its a great tool for describing ontologies (the computer science term for a taxonomy more or less) and for doing so in a way that maps to first order logic and set theory and that can detect inconsistencies in a model that a human might miss. I’ve been using it for some work I’m doing and its really amazing, the first tool I’ve ever seen that really helps you formally analyze and refine an ontology.

  1. I have my Advanced students construct a table with all the animal phyla across the top from poriferans to chordata. Then, down the side we place specific categories. Things like protostome vs. deuterostome or pattern of cleavage or number of embryonic tissue layers… and when the kids step away and look at the patterns they see, they smile and nod and “get it”.

    Taxonomy is very very important in establishing understanding of evolution and the evolutionary advances that accrue along the fossil record.

    We then do the same thing within chordata and look at the different classes of chordates, once again putting strategic categories down the side of the chart. They then see that “wow, this occurs from top to bottom!” We also employ dichotomous sorting to demonstrate the process of classification.

    Richard’s wonderful tome, “The Ancestor’s Tale” is a tour de force of this type of knowledge. I just cannot get over the detail present in our understanding and the richness of the planet that is casually dismissed by people who choose to be ignorant of the study of Biology.

  2. I would have thought if could sequence the DNA of your candidate, you could cook up some sort of algorithm to decide if it was close enough to some existing species to interbreed.

    The categorisation could be mechanised, with sometimes adjustments needed to the tree. Then all you have to do is decide the species name.

    Is the problem not enough species have been sequenced, or even with a sequence of relatives you still don’t know enough?

    • In reply to #5 by Roedy:

      I would have thought if could sequence the DNA of your candidate, you could cook up some sort of algorithm to decide if it was close enough to some existing species to interbreed.

      You get subspecies and ring species with the fertility levels of crosses being a continuous progression. It is only where too big a geographical, time-period, or genetic gap has separated the species gene pools too much to breed, that there is a clear cut boundary between species. Even then you can have infertile hybrids up to a certain level of separation.

      Horses and donkeys produce sterile mules, but some wolves dogs and coyotes can easily interbreed Canid hybrids, as can Polar Bears and Grisly Bears. Ursid hybrids

Leave a Reply