ISPN

[dinoguy2]

So, I was reading this paper (http://palaeo.gly.bris.ac.uk/donoghue/PDFs/2005/Donoghue_2005b.pdf) about crowns, stems, etc., and I got into some dangerous philosophical thinking that was messing with my mind a little, and I figured it'd help to write out my thought process.

A crown group, as we all know, is the clade containing the common ancestor of two living species. What got me thinking was the mention of things like tunicates, and how some researchers were terming them "stem chordates" and such. This got me thinknig about how crown groups are formulated--really, at least among amniotes, all the crown groups correspond almost exactly to the contents of the traditional linnean classes.

For instance, crown group Aves. Why were Passer + Struthio chosen as specifiers rather than, say, Passer + Crocodilus? Obviuosly, because crocodiles are tradionally "reptiles" and ostriches are traditionally "birds". Essentially, defining any group based on extant species is just as arbitrary as Linnean taxonomy becasue it's doing the exat same thing, with a veneer of phylogenetic nomenclature slopped over it.

The telling implication here also is that vernacular terms are directly linked to clade names, especially for crown groups. Why does "bird" = Aves? (Because "aves" *means* "bird"!) Better yet, why was the loaded name Aves even *used* in phylogenetic taxonomy?

If, for instnace, an animal is found to fall within the clade Chordata, that lacks a notochord... well, it wouldn't be the first time the name of a group istelf was misleading as to that group's contents, but still.

I think this whole PN, cladistics, vs oldschool Linnean situation we have now is that much more messy because of the actual nomenclature involved. PN should never have adopted the names of former Linnean groups. Names should be descriptive, especially in PN and cladistics. What's the utility of naming a clade Aves, which implies a very specific suite of apomorphies right in then ame, instead of something like "Passerarchaeopteryx"? Or simply "CrownPasserStruthio" or "StemPasserArchaeopteryx"?

Why not limit the traditional linnean names to something that more accurately refelcts their nomenclature, such as apomorphies? Already we name node- and stem-based clades that may or may not be redundant with each other. Why not have Aves = "Primary feather w/ a central vane as in Passer", Chordate = "Notochord as in Homo", etc. Cladistic methodology, and PN in general, seems always bogged down because it's trying to teach an old Linnean dog new tricks. Why not leave the old names, or at least modify them with definitions matching the contents they describe, and use new names for a new system?

Would certainly fix problems regarding "standard" endings like -idae, at least.

[tmkeesey]

[dinoguy2]

[tmkeesey]

[dinoguy2]

[quote="tmkeesey"]

[tmkeesey]

Etymology does not and should not drive nomenclature. Dinosauria and Basilosaurus are not lizards, synapsids do not have fused fenestrae derived from the diapsid condition, and Pongo pygmaeus is actually quite large.

Of course we can make some concessions to etymology, but we shouldn't bend over backwards. Equating Mammalia with lactation would be bending over backwards. It would make the clade extremely unstable, perhaps nearly as inclusive as Theropsida, perhaps nearly as exclusive as the crown group.

Unstable apomorphy-based definitions may have their place, but not, I think, for such common, well-known, often-used names as Mammalia.

I do agree that there could be a notational prefix for crown groups; perhaps you could say that mammals are corono-therapsids, or that Mammalia = crown(Therapsida).
_________________
T. Michael Keesey
Director of Technology, Exopolis, Inc.

[tmkeesey]

[dinoguy2]

TMk said in the same discussion on another board:

[tmkeesey]

[David Marjanović]

Hey, dinoguy2! Good to see you here! That's easier than discussing on a Wikipedia talk page

I think that apomorphy-based definitions should be used for rather few clades -- those with apomorphies for which we don't need to expect lots of transitional states, and those for which anything but an apomorphy-based definition could result in drastic changes in concept. The turtles (Testudinata sensu Joyce, Parham & Gauthier 2004) fulfill the first criterion: the shoulder blades can be either inside or outside the ribcage. (At least for amniotes, which always have long ribs.) Diapsida fulfills the second: if it gets a branch-based definition, we can discover anapsid diapsids ( = diapsids with an anapsid skull roof), and if it gets a node-based one, we can discover diapsid non-diapsids. Diapsida does not quite fulfill the first condition, though, so a careful choice and delimitation of the apomorphy is required here. I suggest the upper temporal fenestra -- in the right place (ventrally delimited by postorbital + squamosal), larger than a foramen, and with rounded, smooth edges unlike a fontanelle (Milleretta has a lower temporal fontanelle, for example). Tetrapoda may fulfill both conditions, especially if the digital arch hypothesis is correct. (It clearly fulfills the first, in any case!)

In general, however, I don't recommend apomorphy-based definitions. Tetrapoda, for which I still recommend an apomorphy-based definition because of all the above, will illustrate the principal problem: We don't know if Ventastega, Elginerpeton, and several other taxa had digits and/or fin rays, so we don't know if they are tetrapods, even though their phylogenetic position is fully resolved! Anchoring Mammalia on lactation, for example, would be downright foolish.

For Mammalia the presence of the secondary jaw joint (which is not correlated to the absence of the primary one!) could still be used, but this makes Haramiyidae, Tritylodontidae, and Tritheledontidae mammals (they were usually excluded). A node-based definition, such as the one anchored on Sinoconodon (Luo, Cifelli and Kielan-Jaworowska 2002) or the crown-group one, may be a better idea.

For Aves? Forget it.

For Amniota? Anything but the amniotic egg. Eggs fossilize very, very rarely (unless they have a hard shell -- but that's not the normal condition). We don't know how Diadectomorpha and Solenodonsaurus or the possibly relevant Westlothiana and Casineria reproduced, so we wouldn't be able to tell if they are amniotes! Everyone uses Amniota for the crown-group, so let it be that crown-group.

Addition after preview: This forum does not even take en dashes. There are outright Unicode fora out there...

[dinoguy2]

[tmkeesey]

[tmkeesey]

[David Marjanović]

Is there more than one font in this forum? That would explain it very easily.

I wouldn't say that Aves refers to a suite of apomorphies. It's just the Latin word for "birds", and the definition of that is "I know it when I see it". You can try to come up with a phylogenetic definition that keeps the content of this largely intact; I have proposed such a definition on both meetings and hope to publish it sometime. But I can't think of even just one apomorphy that would reliably diagnose such a clade.

[tmkeesey]