NOT-inflated Family Staphylinidae

Michael A. Ivie mivie at MONTANA.EDU
Sat Feb 12 11:17:57 CST 2005

Dear Don,

Since one of the assertions is mine, I will take a stab at this.  However, I will modify your usage to use “monophyletic classification” rather than “cladistic classification” for reasons explained below. The case below will seem simplistic to you, and at first not very satisfying, but it is aimed at those you feel don’t care, not at you, who already know all of this, but may not have thought about it in this form.

My examples will be based on the perfect-case scenario, one we admittedly seldom see, but one that is the goal of every phylogenetic analysis and classification construction.  In this perfect world, a true phylogeny (i.e. the actual line of descent and relationship) is represented by taxa that include the most recent common ancestor, and all descendents of that common ancestor.  Not all biological attributes are known for all members, and new members may be discovered.  However, if one works their way up
the tree, placing attributes on the branches where they are known, the attributes can be read as a series of directional changes, and thus allow predictions of the unknown attributes of  species of interest.

This is because each node represents the addition of a heritable evolutionary novelty that is passed on to descendents, and so a member of the next higher group can be predicted to have certain attributes, even if those attributes are unknown for that individual taxon.  For instance, if a larva of a new Diptera is discovered, it can be predicted to have holometabolous development, because the most recent common ancestor of the Diptera had holometabolous development.  The adult can be predicted to have
specific types of wings, halters, antennae, mouthparts, and on and on because the ancestor of the Diptera had those attributes.   This seems ridiculously obvious, but remember that the evolutionary-based concept of our Diptera was not always the one used.  Under the non-monophyletic classification of Linneaus, this maggot would have been placed in the taxon Vermes.  The only thing that you could predict about it from that classification is that it was worm-like.   It was in a group with nematodes,
earthworms and anything else worm-shaped.  It allowed for a name to be put on everything, but not many predictions to be made from those names.

If this previously unknown dipteran larva has the apomorphic characteristics of the monophyletic Calypteratae, much more exacting predictions about its biology can be made than if it is a member of the non-monophyletic Nematocera.  Of course, it is always possible for the predictions to be wrong – it is really a matter of improving the odds.  If our larva turns out to belong to a wingless species, the prediction about the type of wings will not hold, but the options have been limited.  You will still be
correct that it does not have metathoracic flight wings, for instance.  Thus, with a monophyletic classification, using that classification to predict something currently unknown about a taxon has far better odds of being right than without one.  The fact that many traditional classifications that do not use phylogenetics are able to give good productivity is because they themselves are monophyletic, either without stating monophyly as a goal or not providing a method of evaluating it if they did.  Thus, it
is important to understand that monophyletic taxa are more predictive than non-monophyletic ones irregardless of how the classification was arrived at.  A badly done cladistic analysis that results in a paraphyletic classification is less predictive than an evolutionary taxonomist’s luckily monophyletic one.

I work in a Department dominated by applied entomologists, and I understand your assertion that they do not seem to care about the monophyly of a taxon.  What they really do not care about, in my experience, is the details of the distinction.  When confronted with a discussion of monophyly vs non-monophyly, their eyes glaze over and they get a faraway look or seek something with which to commit suicide.  They just want us to do our job, and deliver the classification, not explain its details.   I find that
in the main, users of our classification assume that we have given them the predictive power of a monophyletic classification, and go on that assumption when trying to make predictions.  If a species group has several members that transmit a given disease, they assume it is more likely that other members of that species group will transmit the disease than members of another species group.  If all members of a genus whose biology is know feed only on Euphorbiaceae, they assume that the new one they found
will feed on Euphorbiaceae.  This is in fact the conclusion that would derive from a monophyletic classification, but not from a para or polyphyletic classification.  They figure we have been at this of 250 years, we should have gotten it right.  Little do they know (or care) the truth.

As a practical example, I remember an analysis by an Australian of a group of  Sternorrhycha (part of what used to  be the Homoptera, but I don’t remember which one) that was a terrible problem in Oz.  Biological control efforts had been aimed at Asia, where most of the pestiferous members of this very large and widespread genus occurred, and from where the Australian populations were thought to be from.  None of the parasitoids found would touch it, so a phylogenetic analysis was done of the whole genus,
and low and behold, the Australian pest ended up in a clade of temperate South American species, but was none of the species actually known from there.  Biocontrol exploration was sent to the region, and found conspecifics and good biocontrol candidates within a few days.  That is the type of predictability that only a cladistic analysis and a monophyletic classification can give, and the type of example that our applied colleagues can use.


Don.Colless at CSIRO.AU wrote:

> Dear All,
> We have lately had several assertions to the effect that a cladistic classification is the "most predictive". May I ask just what it predicts, apart from the likely monophyly of its taxa?
> My problem, perhaps, is that I've worked for much of my life in applied entomology, where folk (non-taxonomists) are interested in predicting, for instance, that any member of the genus Anopheles is a potential vector of malaria; or that any member of the genus Anagonia is almost certainly a parasite of leaf-eating beetle larvae. Even if told that these genera were paraphyletic (and I think the second might be), and even if they knew the meaning of "paraphyletic", I very much doubt that they would care!
> Don Colless,
> Div of Entomology, CSIRO,
> GPO Box 1700,
> Canberra. 2601.
> Email: don.colless at
> Tuz li munz est miens envirun


Michael A. Ivie, Ph.D.
Department of Entomology
Montana State University
Bozeman, MT 59717

(406) 994-4610 (voice)
(406) 994-6029 (FAX)
mivie at

More information about the Taxacom mailing list