taxon as collection of organisms definition is inconsistent with BFO superclass

[cmungall]

taxon as collection of organisms:

A collection of organisms that has as members every organism of a taxon and no organisms of any other equal or higher ranked taxon. The taxon may be at any level. If the taxon is species, then this is the same as the class species as a collection of organisms. When defining a taxon, it is generally understood that members include both living organisms and dead organisms that are known through specimens or observations.

"dead organism" is not defined (is this an actual material corpse, or a referent to a long dead organism that has left no material trace, e.g. the majority of dinosaurs)

However, without being an ontological pedant, the intent of the definition seems clear. The instance "metaozoa as collection of organisms" seems intended to denote a material object aggregate with members:

• me and you
• my pet fish
• terry the T-rex, a dinosaur that has left no fossil trace
• various assorted members of long extinct metazoa

However, this is not consistent with the BFO class of continuant, which this inherits from:

A continuants parts are all present at one time, and member-of is a subrelation of part-of

For this to work you need to define property member-of-or-historic-member-of

Add a property chain:

• descended-from o member_of -> historic-member-of

(if descended-from is locally reflexive you can simply just have a non-proper 'historic-member-of' . See https://oborel.github.io/obo-relations/reflexivity/)

Here descended-from would be a relation that holds between populations/organism aggregates, like develops-from with cell lineages

Thus terry the t-rex is a historic member of the collection "metaozoa as collection of organisms anthropocene", which is (transitively) descended from "metaozoa as collection of organisms jurassic" which has terry as a direct member.

As an aside, I think pco should come with a set of exemplar abox axioms that can simultaneously illustrate the sometimes meta-level thinking in pco, as well as serve as unit tests.

[ramonawalls]

Thanks, Chris. This is really helpful. I will add it to the next release, so as not to delay this one.