In phylogenetics, reticulation is when genes from multiple branches end up in the same daughter lineage. Thinking of it as a family tree, this means that instead of the branches only splitting over time, occasionally they make a network. For example, my ancestors 100K years ago include Homo neaderthalensis as well as Homo sapiens, and possibly also Homo denisova.
This is certainly similar to the Linnaean concept of hybridization, and often results from sexual hybridization, but can also result from less familiar forms of recombination between viruses or bacteria, and even other means of genetic materials moving between relatively unrelated organisms. The endosymbiosis between archaea and bacteria that resulted in eukaryotes is a good example of a very deep reticulation. The reported uptake by tardigrades of genetic material from their food suggests another.
What I’m wonder is if there are any cases, other than present day hybridization, where past reticulations complicate the branching Linnean taxonomy that iNat uses. To give a theoretical example, there is no hard reason except custom why a single genus couldn’t belong in two separate families, if past members of these families (before they had diverged so much) bred and gave rise to the new genus. In practice most taxonomists would likely elevate this hybrid genus to a family along side the other two, or collapse the whole mess into a single family.
But I’m curious if there are any case where taxonomist have seen fit to recognize reticulation, at the generic level or higher, in ways that iNat either has to treat or purposefully sidestep? And are there groups (viruses, perhaps) where reticulation happens so often that the assumption of a tree-like taxonomy has to be largely abandoned? Thank you.
None immediately come to mind, but I would clarify that iNat does not create its own taxonomic system or make taxonomic judgements, just implements the taxonomic system of the relevant authority for the taxa. If there were a case like this, that relevant authority would have to make whatever changes to their system and iNat would follow suit.
I agree that iNat follows the relevant taxonomic authorities, but deciding which authorities to follow, and how, and when, is a set of taxonomic choices. No where else in the world will you find exactly iNat’s taxonomy, because no one else is making exactly the same choices.
I think @intyrely_eco’s point is a valid one… while choices might have been made about what authority to follow, that decision likely doesn’t weight heavily on how reticulation is implemented by them. Having said that, @loarie might be aware of any impacts related to reticulation with regards to any past importation of taxonomic structures from any of the authorities…
It strikes me that the taxa that are most likely to be affected by this reticulation are the ones that are least likely to be identified in iNat at a fine enough level to experience it. We are largely talking about field identifiable and/or photographic evidence, and only rarely do we see micro work or dissection etc. The “placement” of taxa in iNat usually gets attention when those that identify the taxa encounter the literature associated with changes, and the only taxa literature being monitored tends to be those that have observations made of them… ie there aren’t a lot of volunteers working on parts of the system that never get used…
Perhaps one approach to this question would be to establish several taxa that it does come into play on, and then look at iNat taxonomy/observations of them to see if they are affected.
Yes, there are very few people making numerous observations of viruses, for example. But those viruses that do get observed (often because of their visible and identifiable effects on larger organisms) are where this seems most likely to come up.
As far as I know, iNat’s taxonomy is strictly a tree structure, so it cannot represent such net structures. Technically - if a genus for example cannot be assigned to a (single) family - it can be assigned to a higher taxon as incertae sedis and thus “skip” the reticulate portion of the phylogenesis. For iNat’s purposes, representing such intricate reticulate relationships is not needed.
Most of iNat’s taxonomy is implemented by mapping high level taxa to external taxonomy providers such as Catalogue of Life. External taxonomy providers also use tree representations so iNat taxonomy curators probably never deal with problems of reticulation. You can read more about how iNat’s taxonomy is built here: https://www.inaturalist.org/pages/taxon_frameworks