[Ultimatt]

I dunno you can mostly solve for this, look at the majority share of the genome in a compartment with a single membrane contained by the outer membrane (by definition for endosymbiogenesis), with the DNA molecules containing the highest quantity of coding gene content. That's the thing you do the cladistic descent on. Just because I'm carrying a load of bacterial symbionts that are better at digesting wheat doesn't make me a different species to humans who mostly eat rice. That I eventually carry those as an endosymbiont then eventually just a sack of genes doesn't really change who' was the boss in this situation, or who I'm able to reproduce with that entire time as a species either. Symbiogenesis is very much like this, look at Hatena arenicola if you want to see this emerging to see how it interacts with descent. This is really the difference between understanding reproductive relationships specifically, vs wider ecology. I agree that symbiogenesis is like this weird one where its both only going to happen through other ecological relationships and then reproduction. But there really is a clear cut time one turns into the other, with the fate of one of those organisms of the combination "winning" even though profoundly altered. Symbiodinium is probably another good example, it has loads of ecological partners and little remnants of all of those histories in its genetic material and organelle arrangement. But its still an obvious species, with all the other oddball dinoflagellates that act and look like it also appearing to have some common descent from something much simpler and eukaryotic.

As a programmer symbiogenesis is a mixin or interface not a class.

[adrian_b]

It is not that simple, because a large part of the genes that originally belonged to the mitochondria or to the chloroplasts have been transferred into the nucleus and they are now completely integrated into it, even if evolution trees will show that those genes have a different history than the remainder of the nucleus.

So even if you want to use only the DNA inside the nucleus for a cladistic classification, that does not produce a tree of evolution, but only a directed graph of genetic information flows, with many important hybridization events. While initially mitochondria and chloroplasts were just symbionts, like the useful bacteria in the human gut, eventually that symbiosis has become a full hybridization, with mixing and integration of the genetic information.

The official cladistic classification produces a tree from the directed graph of the evolution by cutting the branches that are considered to be less important.

Sometimes this is indeed the best choice, but there are contexts in which the genetic information brought through the minor branches is actually that which determines most of the importance of an organism in an ecosystem.

E.g. in many contexts the most important feature of a living being is whether it is an oxygenic phototroph due to inheriting genetic information from some blue-green algae, i.e. that it is a primary producer in the ecosystem, and not the features that depend on which is the exact eukaryotic group from which most of its nucleus has been inherited. For instance, it is more frequently useful to group together brown algae with red algae (whose chloroplasts share a common ancestor), than to group brown algae with some non-phototrophic stramenopiles (which share a common ancestor for most of the nuclear DNA), with which they share only inherited characters that need an electron microscope for detection.