The integrative future of taxonomy

[racoll]

Really nice introductory review on modern taxonomic practise.

L numbers even get a mention!

Tons of must-have references too.

http://www.frontiersinzoology.com/conte ... 4-7-16.pdf

[Mike_Noren]

It's an interesting paper, but I'm a bit puzzled as to what it's point is.

The "integrative taxonomy by cumulation" has one single necessary component: "taxonomist considers observed differences in one or several characters to be taxonomically relevant". Everything else is optional. Yes, nearly all species are of this kind, but it is still subjective pre-cladistic inductive thinking, I really think taxonomy has or at least should move past that now.

The "integrative taxonomy by congruence" equals the criterion of diagnosability except it has to be observable in more than one data source (why?).

To me diagnosability is king. Are the groups of animals diagnosable? Fine, that proves reproductive distinctiveness, so consider them separate species. Not diagnosable? That proves significant interbreeding, so don't consider them different species. I don't see the point in making it more complex than that.

[Silurus]

[Moderator's note] - I have moved this out of the Taxonomy forum because I am keeping it strictly catfish-specific.

[racoll]

[Moderator's note] - I have moved this out of the Taxonomy forum because I am keeping it strictly catfish-specific.

Fair enough.

What happened to my other post on Brazilian fish conservation by the way? Did I do something wrong? It seems to have been deleted rather than moved.

The "integrative taxonomy by cumulation" has one single necessary component: "taxonomist considers observed differences in one or several characters to be taxonomically relevant". Everything else is optional. Yes, nearly all species are of this kind, but it is still subjective pre-cladistic inductive thinking, I really think taxonomy has or at least should move past that now.

I think "taxonomy by cumulation" is not necessarily "pre-cladistic" as it works with the criterion of diagnosability too, it's just that it relies on a single character to differentiate groups rather than several (i.e. congruence).

[MatsP]

Racoll, Sssh. Don't tell anyone - Silurus probably made a mistake, and deleted the post. At least that's what the moderator logs say...

--
Mats