3 ms·

I mean, dolphins are fish, in the sense that the most recent common ancestor of all modern fish is also the common ancestor of all modern dolphins. Very few ani

I mean, dolphins are fish, in the sense that the most recent common ancestor of all modern fish is also the common ancestor of all modern dolphins. Very few animals aren’t fish in this sense though. Birds are also fish. What’s surprising is that you can exclude sharks and rays from your definition of ‘fish’ and this statement will still be true - you are more closely related to a salmon than a salmon is to a shark.

goldenkey 6y ago
It's not just a rooted tree, it has crossovers and cycles. Trying to draw boundary boxes is a fools hardy.
- jameshart 6y ago
  This is true, when you zoom in closely, but when you zoom out, it is quite treelike. phylogeny isn't useless just because the process of speciation isn't instantaneous.
  - User23 6y ago
    It’s very helpful to learn to recognize lumper versus splitter arguments. Pointing out that lumping things together or splitting things apart is arbitrary just isn’t very interesting. What is interesting is why some things are lumped and some are split. For example why are polar bears and grizzly bears considered different species? They can breed just fine after all.
    - jameshart 6y ago
      Right, and when you notice that a bunch of organisms share a trait, the fact that they share that trait in spite of their common ancestor not sharing it is inherently interesting, right? We know that traits are heritable, so when you have a bunch of different organisms that exhibit the same characteristic, if you can trace back and determine they are all specialized descendants of some ancestor which first developed that characteristic, that’s the complete story right there. But when we see the same behavior appear in different organisms without them sharing a common root for that characteristic that’s interesting precisely because we need another means to explain it. I mean, if grizzlies and polar bears could interbreed even though their nearest common ancestor was, say, some kind of small Cretaceous rodent, then weld have some explaining to do, right?
      - User23 6y ago
        Absolutely. I find the cladistics approach appealing. On a bit of a tangent I did some searching and found this[1]. Nightjars are cool! [1] https://scienceblogs.com/gregladen/2010/09/09/one-could-say-that-bats-evolve https://scienceblogs.com/gregladen/2010/09/09/one-could-say-...

goldenkey 6y ago

It's not just a rooted tree, it has crossovers and cycles. Trying to draw boundary boxes is a fools hardy.

jameshart 6y ago
This is true, when you zoom in closely, but when you zoom out, it is quite treelike. phylogeny isn't useless just because the process of speciation isn't instantaneous.
- User23 6y ago
  It’s very helpful to learn to recognize lumper versus splitter arguments. Pointing out that lumping things together or splitting things apart is arbitrary just isn’t very interesting. What is interesting is why some things are lumped and some are split. For example why are polar bears and grizzly bears considered different species? They can breed just fine after all.
  - jameshart 6y ago
    Right, and when you notice that a bunch of organisms share a trait, the fact that they share that trait in spite of their common ancestor not sharing it is inherently interesting, right? We know that traits are heritable, so when you have a bunch of different organisms that exhibit the same characteristic, if you can trace back and determine they are all specialized descendants of some ancestor which first developed that characteristic, that’s the complete story right there. But when we see the same behavior appear in different organisms without them sharing a common root for that characteristic that’s interesting precisely because we need another means to explain it. I mean, if grizzlies and polar bears could interbreed even though their nearest common ancestor was, say, some kind of small Cretaceous rodent, then weld have some explaining to do, right?
    - User23 6y ago
      Absolutely. I find the cladistics approach appealing. On a bit of a tangent I did some searching and found this[1]. Nightjars are cool! [1] https://scienceblogs.com/gregladen/2010/09/09/one-could-say-that-bats-evolve https://scienceblogs.com/gregladen/2010/09/09/one-could-say-...