Nested Biological Hierarchies

Another wonder of the nested hierarchy. The fossilized dolphin would post a major problem if it was thought to be a experimenting fish. Since it would have to have also experimented with a radically different body plan from a fish. Fish spines move side-to-side, rather than the mammalian manner (which all cetaceans follow). Fish don't have fingers or a whole plethora of other features we'd find in a dolphin. It has a whole bunch of characteristics which would mark it as a unique entity from all other fish. Indeed it is only certain 'base' characteristics that it shares: It is alive, an animal and a vertebrate.Nobody would classify them as similar based on any traits other than 'aquatic', which is about as useful as 'land dwelling' as classification goes. We need a classification scheme which is based on unique characteristics. If you want we can list all the characteristics unique to mammals (basic body plan only if you'd like to think in terms of fossils) and all the characteristics unique to fish.Then we can see how to classify our hypothetical dolphin fossil.We can skip this if you''d like though. I can guarantee it will be impossible to classify a dolphin as something other than an animal and a vertebrate and a mammal and a cetacean. The proof is in the pudding if you want to try it out.If you want to try defining 'fish' as 'lives in water' we run into classification problems since somethings live slightly in water, some live mostly in water. Indeed, dolphins live in water, but breathe out of it...indeed, classifying animals based on subjective things such as behaviour is frought with danger. We should look to more objective things such as body form, genetics etc.Its like classifying a library. Is the Italian cookbook written in French something we should classify as an Italian culture book, a recipe book or a book written in French?That's why we look for distinguishing characters. Books don't fall into a unique nested hierarchy, but life does.

---

But here is the great thing - we can do the nested hierarchy thing with absolutely no fossils whatsoever! We can do it with extant lifeforms and get the same result. And still we'd find ourselves amazed by it.

---

Edit: The actual title of the piece is '29+ Evidences for Macroevolution', and is not intended to be exhaustive. It also doesn't count each tiny piece of evidence towards that total, but general evidences. For example, it doesn't mention every single type of atavism, just some of the more common/obvious ones. Just a random AbE really, an fyi you might say. No need to worry too much about it.Edited by Modulous, : No reason given.

It is a good point. We could make hierarchies from any collection of things and it would be entirely subjective. The thing that makes the biological hierarchy so compelling is that it is (in Theobald's words) 'a unique, consistent, well-supported tree that displays nested hierarchies.', he goes on to say Which I think sums it up. Right after this he goes on to discuss statistical methods for calculating the subjectivity or objectivity of a given hierarchy. Sure enough - biological hierarchy gives a high result (objective) whereas arbitrary hierarchies give low results (subjective). It discusses one method in the potential falsification section.So the maths has to be addressed. Maybe you think the maths is bogus, which would mean explaning why, or perhaps you agree with the maths but disagree with what it means...once again stating why.Another interesting thing about nested hierarchies is that we don't find new species which don't fit neatly in one place within it. There are times when classification is difficult, where both phylogenies are equally well supported. That said, these difficulties are generally about more closely related species and not about wider ranging things; we don't find a species where we're not sure if it is a mammal or a fish for example.The final difficulty I've not seen you really address is that morphological studies can be used to create a unique hierarchy and so can genetic traits (even traits that do not dictate morphology). The two unique hierarchies generated are very very similar, sometimes identical.The statistical probability of this happening is vanishingly small. So eithera) DNA structures that do not affect an organisms physical traits are somehow and in some significant manner related to its physical traits. For example - great apes are morphologically close. But they share very very similar genetic sequences for cytochrome c. Cytochrome c is not related to the organisms morphology. Indeed human cytochrome c functions in yeast, as do many other types of cytochrome c.b) This kind of congruence was the result of intentional intervention by some unknown entity.c) A gigantic coincidence.The first answer, a) raises a question (What's the connection?), the answer is heredity which has been observed (organisms do replicate). The second answer b) raises another question (what are this being's motivations for doing this?), which is often theologically difficult for those inclined to insert their deity of choice here. c) is absurd.

I never mentioned transitions, I was talking about classification. The two can be related, but they are essentially seperate topics. Well, let's use the hierarchy to explore this issue. First the Hyrax: (source)And the manatee: (source)Using the nested hierarchy of life alone I can predict (I don't know at this point) that hyraxes and manatees are quite distantly related. They share the same superorder - Paenungulata. I go away and look up this superorder to find out; we're looking at the most recent common ancestor between these two things being around at the same time as the dinosaurs! The nested hierarchy was right, pretty distantly related really.So why would you bring it up? You probably heard that the closest living species to hyraxes are manatees, am I right? The reason is that the other relatives are all dead. If we were the only primates, we'd have a similar dilemma when discussing our closest relative. With 65million years and a generational time of ten years, we are talking about 6.5 million generations. Its probably more than that. Nobody is saying that they are sure of every single one of them. There is considerable dispute indeed, on the exact details. If we were to make it into a film and watch every generation at 100 generations per second (television is only 30 frames per second for comparison) it would take 18 hours to watch the whole thing. We would miss most of the frames (though each frame would be pretty much identical to the last one). This is using a very long generational time of ten years. It is most likely to be much lower than that.Hyraxes reach sexual maturity at about 18 months. If we assume their generational time is about 2 years. Manatees are a little more complicated - about 5-9 years. So if we reduce the generational time to their average we get about 6 years or so, which would increase our film length to 30 straight hours of frames, each nearly identical to the one before.(Assuming total gradualism which is not what we observe, this is just to impress upon you the scale of change from one generation to the next that we are talking about here). Unless the fossil record of our extinct dolphins was particularly bad, we should be able to classify it quite easily. If we had access to its ear we would immediately classify it as a cetacean. Pretty much any bones would enable us to classify it as a mammal, if things were pretty bad we'd probably get stuck on tetrapod. It would have to be the most direst collection of bones to not be sure if it was a fish or a mammal.Edited by Modulous, : Adding the film metaphor.

It takes a while to do sequence comparisons, but here are the sequences: silkwormlampreyturtlePigeonHorseand the result of the comparison:Sequence 1: silkworm 107 aa
Sequence 2: lamprey 103 aa
Sequence 3: turtle 104 aa
Sequence 4: pigeon 104 aa
Sequence 5: horse 104 aaSequences (1:2) Aligned. Score: 70.8738
Sequences (1:3) Aligned. Score: 76.9231
Sequences (1:4) Aligned. Score: 77.8846
Sequences (1:5) Aligned. Score: 75.9615
Sequences (2:2) Aligned. Score: 100
Sequences (2:3) Aligned. Score: 79.6117
Sequences (2:4) Aligned. Score: 80.5825
Sequences (2:5) Aligned. Score: 82.5243
Sequences (3:2) Aligned. Score: 79.6117
Sequences (3:3) Aligned. Score: 100
Sequences (3:4) Aligned. Score: 92.3077
Sequences (3:5) Aligned. Score: 89.4231
Sequences (4:2) Aligned. Score: 80.5825
Sequences (4:3) Aligned. Score: 92.3077
Sequences (4:4) Aligned. Score: 100
Sequences (4:5) Aligned. Score: 89.4231
Sequences (5:2) Aligned. Score: 82.5243
Sequences (5:3) Aligned. Score: 89.4231
Sequences (5:4) Aligned. Score: 89.4231
Sequences (5:5) Aligned. Score: 100Not easy to read, so I can make a N-J tree: This is the way we should do these things. The interesting thing here is that the pigeon and turtle are grouped together, as per the hierarchy. You will note that it isn't perfect, which leads to my other point:Just testing one protein is dangerous and firm conclusions cannot be drawn, but general tendencies are useful (and such a small number of organisms doesn't help, I don't have the time now, but might later to expand on this). One thing to bare in mind is that different techniques need to be applied with non-vertebrates, so a straight comparison between vertebrates and non-vertebrates is also fraught with potential error. There are some interesting resources out there, but they are a topic in themselves, if you are interested let me know.

It takes a while to do sequence comparisons, but here are the sequences: silkwormlampreyturtlePigeonHorseand the result of the comparison:Sequence 1: silkworm 107 aa
Sequence 2: lamprey 103 aa
Sequence 3: turtle 104 aa
Sequence 4: pigeon 104 aa
Sequence 5: horse 104 aaSequences (1:2) Aligned. Score: 70.8738
Sequences (1:3) Aligned. Score: 76.9231
Sequences (1:4) Aligned. Score: 77.8846
Sequences (1:5) Aligned. Score: 75.9615
Sequences (2:2) Aligned. Score: 100
Sequences (2:3) Aligned. Score: 79.6117
Sequences (2:4) Aligned. Score: 80.5825
Sequences (2:5) Aligned. Score: 82.5243
Sequences (3:2) Aligned. Score: 79.6117
Sequences (3:3) Aligned. Score: 100
Sequences (3:4) Aligned. Score: 92.3077
Sequences (3:5) Aligned. Score: 89.4231
Sequences (4:2) Aligned. Score: 80.5825
Sequences (4:3) Aligned. Score: 92.3077
Sequences (4:4) Aligned. Score: 100
Sequences (4:5) Aligned. Score: 89.4231
Sequences (5:2) Aligned. Score: 82.5243
Sequences (5:3) Aligned. Score: 89.4231
Sequences (5:4) Aligned. Score: 89.4231
Sequences (5:5) Aligned. Score: 100Not easy to read, so I can make a N-J tree: This is the way we should do these things. The interesting thing here is that the pigeon and turtle are grouped together, as per the hierarchy. You will note that it isn't perfect, which leads to my other point:Just testing one protein is dangerous and firm conclusions cannot be drawn, but general tendencies are useful (and such a small number of organisms doesn't help, I don't have the time now, but might later to expand on this). One thing to bare in mind is that different techniques need to be applied with non-vertebrates, so a straight comparison between vertebrates and non-vertebrates is also fraught with potential error. There are some interesting resources out there, but they are a topic in themselves, if you are interested let me know.