Taxonomy as a Rigorous Science

[Jodie]

I've never used statistics as a point of discussion in my life. I don't understand it, I'm not educated in that field of study. There's no shame in conceding one's limitations.

Then how in the hell can you ever read any kind of research and understand whether the conclusions are even close to being sound? How can you judge anyone's ability to read research and know whether any kind of critical thinking skills were used to assess the conclusions? Obviously you can't.

 

[Dinwar]

Can I just jump in, as a professional taxonomist...

Of course. I was sort of hoping you would. I know that you're far more well-versed in taxonomy than I am, and would appreciate any criticism you have.

I know that my description isn't entirely accurate according to current usage. However, it's what I learned. I purposefully found an old-school person to teach me taxonomy, since my undergrad work was very heavily statistically based.

No it doesn't contradict the meaning of "subjective". How many times do I hear this same criticism applied to other research where the data is incomplete or variables aren't known.

Depends on who you listen to. If you listen to legitimate scientists, not a single one. There is a world of difference between subjective and not supported well enough for you, and your failure to learn the fields you choose to attack in no way eliminates that difference.

Exactly, that was my point.

You've officially stopped making sense.

No Dinwar, I accept the limitations, I think we are looking at this from two different directions.

Well, you're right about that list phrase anyway. I'm looking at it from the perspective of someone who's educated themselves on the issues, problems, and current methods of the field. You demonstrably cannot identify the fields involved.

You are wrong, I provided a mathematical equation that demonstrated what number of fossils would be necessary to have to accurately represent the variation in a Neanderthal population. It doesn't get any more concrete than that.

One which was, as I recall, roundly criticized and based on assumptions that make the typical Creationist rant look reasonable.

And if you think that merely having an equation makes something non-subjective, I suggest going to whatever university you attended and demanding your money back. You got screwed. Merely having an equation DOES NOT make something objective. It merely paints a facade of objectivity over it. You say you studied statistics--did you learn NOTHING in your studies? Any freshman in a stats class can use math to prove nearly anything they want.

I'm not going to continue to refute your points. The fact that you believe merely having an equation makes your side objective, while the other side is subjective, states more eloquently your failure to understand the concepts involved than anything I could. It also further illustrates your ignorance of hte topic, as these topics have been studied by paleontologists for well over 200 years.

 

[Jodie]

Well you just undermined your own assertion that your objective methods were sound, now you aren't making sense.

 

[Vortigern99]

Then how in the hell can you ever read any kind of research and understand whether the conclusions are even close to being sound? How can you judge anyone's ability to read research and know whether any kind of critical thinking skills were used to assess the conclusions? Obviously you can't.

"How in the hell" is that I understand the scientific method, and I have undergrad training in the collection and presentation of data. I've made it clear throughout my tenure here at the JREF that I'm an anthropological hobbyist, not a professional scientist or academician. Statistical expertise is not required to grasp even the tiniest details of human phylogeny, nor is your pet field of mathematics needed to follow comparative ethnography. Those are my areas of interest and I can speak and write about them with authority.

Critical thinking is a system of comparative data analysis; it doesn't depend on any extensive knowledge of statistics, despite what you appear so passionately to believe. I find it remarkable that you place such primacy on the one area in which you appear to be well-educated, and that you judge others' intellectual capacities based on our familiarity with that single subject.

I am absolutely finished discussing anything with you, as I find your anger unpleasant and your logic muddled. Enjoy your delusions!

 

[Jodie]

How do you determine whether the data is sound if you don't have a basic grasp of statistics? Take the researcher's word for it?

Fine then, I see I hit a nerve on that one.

 

[Vortigern99]

Yes, here at the JREF it's all about "hitting a nerve" and scoring points! Never mind the thread subject, let's find out what Poster X doesn't know about and poke him there for our own amusement!

I've answered your question in my previous post. Did you read it?:

I understand the scientific method, and I have undergrad training in the collection and presentation of data. Statistical expertise is not required to grasp even the tiniest details of human phylogeny, nor is your pet field of mathematics needed to follow comparative ethnography. Those are my areas of interest and I can speak and write about them with authority.

Critical thinking is a system of comparative data analysis; it doesn't depend on any extensive knowledge of statistics, despite what you appear so passionately to believe. I find it remarkable that you place such primacy on the one area in which you appear to be well-educated, and that you judge others' intellectual capacities based on our familiarity with that single subject.

Address that, or pipe down.

 

[Dinwar]

How do you determine whether the data is sound if you don't have a basic grasp of statistics? Take the researcher's word for it?

You learn the field. If you refuse to do that, than yes, you are obliged to take the researchers' word on the topic. Please note that I'm not saying anything about you--I always point out at this point that for me, electricity falls firmly into this category. I don't know it, so I leave it to the experts to figure it out. I use what they determine, as best as my limited knowledge can determine.

As for the requirement to have a collection of a certain size before you can say something is a new species, there are a few flaws in this.

First, there's the fact that there are statistical methods for determining the probability of a sample falling within a known population. In fact, that's what standard deviations are often used for--if you are more than two or three standard deviations from the mean, there's a very high likelyhood that you are not a member of that population. Since we have LOTS of human bones, we can calculate with some degree of confidence whether or not an individual specimen is H. sapiens sapiens or not. Thus, statistical analysis disagrees with your conclusion.

But I'm still willing to entertain your idea. In order to demonstrate its validity, please list the assumptions your equation makes when applied to paleontological data.

Second, the concept of "species" is an issue. The precise place where you divide two species has always been fuzzy. We can determine the range it must fall in, but by necessity any two related species are extremely similar to each other during and immediately after speciation. Speciation isn't typically allopatric; in fact, I'd say it's more akin to ring species. One species develops numerous subpopulations, which diverge from one another over time. At some point, something happens that removes the intermediate forms (remember, at the time that event occurs they are still considered one species); after the event, the populations become their own species. OF COURSE we will find intermediates showing that at one point all these species were one; that's trivially obvious to anyone with any knowledge of evolution. This doesn't change the fact that AFTER the event they are different species. Normally this isn't an issue; speciation happens so fast that we generally only get a few intermediates to deal with. And frankly no one's interested in speciation events in brachiopods or Callianasids or the like. Humans, though, occur extremely late in the fossil record, which means we have a very fine-scale record of them, and they are intensely studied (the mere fact that their study has its own special name is proof of that). That means that we need to address edge issues that are normally irrelevant.

I find it remarkable that you place such primacy on the one area in which you appear to be well-educated, and that you judge others' intellectual capacities based on our familiarity with that single subject.

I don't. I've found that those who have fetishes for particular fields of knowledge tend to judge everyone based on their understanding of those fields, regardless of the validity of that judgement. Sheldon Cooper is extreme, but he's only different in degree, not in type. I've run across physicists that actually do think that anything other than physics is a waste of time. I've also met minerologists who insist that anything else isn't REAL geology ('All rock is made of minerals!"), chemists who insist that ecologists and taxonomists are ignorant because they don't know chemistry, etc. Some people latch onto a single idea, and forget that there's a wider world of knowledge. Unfortunately, statistics tends to either attract such people, or at least provide enough positive feedback that they conclude that in their case, they're right. Try arguing that one can conduct science in ordinary language on this very board sometime. I have--and the results were a rather rabid chorus of "Science requries statistics!!!!"

 

[Jodie]

Yes, here at the JREF it's all about "hitting a nerve" and scoring points! Never mind the thread subject, let's find out what Poster X doesn't know about and poke him there for our own amusement!

I've answered your question in my previous post. Did you read it?:

I understand the scientific method, and I have undergrad training in the collection and presentation of data. Statistical expertise is not required to grasp even the tiniest details of human phylogeny, nor is your pet field of mathematics needed to follow comparative ethnography. Those are my areas of interest and I can speak and write about them with authority.

Critical thinking is a system of comparative data analysis; it doesn't depend on any extensive knowledge of statistics, despite what you appear so passionately to believe. I find it remarkable that you place such primacy on the one area in which you appear to be well-educated, and that you judge others' intellectual capacities based on our familiarity with that single subject.

Address that, or pipe down.

Well you jumped in the conversation and proceeded to criticize what you called my lack of critical thinking. All I'm asking is if you can't judge for yourself whether the algorithms used to establish these systems for phylogeny are sound how can you critically think through any possible hypothesis or theory? It's a legitimate question. I guess it's a matter of being able to dish it out but not being able to accept the same reflective criticism.

 

[Jodie]

You learn the field. If you refuse to do that, than yes, you are obliged to take the researchers' word on the topic. Please note that I'm not saying anything about you--I always point out at this point that for me, electricity falls firmly into this category. I don't know it, so I leave it to the experts to figure it out. I use what they determine, as best as my limited knowledge can determine.

As for the requirement to have a collection of a certain size before you can say something is a new species, there are a few flaws in this.

I was talking about what makes something a variation within a species versus being identified as a new species without the DNA, but yes, you have a point.

First, there's the fact that there are statistical methods for determining the probability of a sample falling within a known population. In fact, that's what standard deviations are often used for--if you are more than two or three standard deviations from the mean, there's a very high likelyhood that you are not a member of that population. Since we have LOTS of human bones, we can calculate with some degree of confidence whether or not an individual specimen is H. sapiens sapiens or not. Thus, statistical analysis disagrees with your conclusion.

Well my thought was that if you didn't have a large enough representative sample to begin with that any statistical analysis you would be using would be off.

But I'm still willing to entertain your idea. In order to demonstrate its validity, please list the assumptions your equation makes when applied to paleontological data.

We started to address that with the Neanderthals at the beginning of the thread but the thread veered off into whales. We assumed, meaning that figure was plucked randomly out of the air, that 10% of the Neanderthal fossils found should indicate some kind of deformity if 10% of the HSS population had that incidence rate. The reason being that Neanderthals are closer to us genetically than chimps, there would probably not be a huge statistical difference. We assume that the total population for Neanderthals was 70,000 with only something like 400 fossils found so far. Is that enough to obtain any kind of idea of what would indicate variation or deformity or would you assume it was a whole other species if you did find something like that?

Second, the concept of "species" is an issue. The precise place where you divide two species has always been fuzzy. We can determine the range it must fall in, but by necessity any two related species are extremely similar to each other during and immediately after speciation. Speciation isn't typically allopatric; in fact, I'd say it's more akin to ring species. One species develops numerous subpopulations, which diverge from one another over time. At some point, something happens that removes the intermediate forms (remember, at the time that event occurs they are still considered one species); after the event, the populations become their own species. OF COURSE we will find intermediates showing that at one point all these species were one; that's trivially obvious to anyone with any knowledge of evolution. This doesn't change the fact that AFTER the event they are different species. Normally this isn't an issue; speciation happens so fast that we generally only get a few intermediates to deal with. And frankly no one's interested in speciation events in brachiopods or Callianasids or the like. Humans, though, occur extremely late in the fossil record, which means we have a very fine-scale record of them, and they are intensely studied (the mere fact that their study has its own special name is proof of that). That means that we need to address edge issues that are normally irrelevant.

My scenario might not be a relevant edge issue in paleontology but it would be in my modern day work world where you are trying to decide which populations and what disease incidence warrants allocation of funds.

I don't. I've found that those who have fetishes for particular fields of knowledge tend to judge everyone based on their understanding of those fields, regardless of the validity of that judgement. Sheldon Cooper is extreme, but he's only different in degree, not in type. I've run across physicists that actually do think that anything other than physics is a waste of time. I've also met minerologists who insist that anything else isn't REAL geology ('All rock is made of minerals!"), chemists who insist that ecologists and taxonomists are ignorant because they don't know chemistry, etc. Some people latch onto a single idea, and forget that there's a wider world of knowledge. Unfortunately, statistics tends to either attract such people, or at least provide enough positive feedback that they conclude that in their case, they're right. Try arguing that one can conduct science in ordinary language on this very board sometime. I have--and the results were a rather rabid chorus of "Science requries statistics!!!!"

I'm not saying what you are doing is wrong, it's the best you can do with what you have to work with, I get that. I'm not saying that you do this, but I do see people here using the subject as if it was concrete fact, when it isn't.

 

[Akri]

Well my thought was that if you didn't have a large enough representative sample to begin with that any statistical analysis you would be using would be off.

The problem is that you haven't explained how large is "large enough", or how you make that determination. According to Dinwar most species have very little morphological variation, and so a huge sample size isn't needed. You've given us no reason to think this is wrong.

We started to address that with the Neanderthals at the beginning of the thread but the thread veered off into whales.

As has been explained multiple times, that paper is directly related to the topic at hand. Saying that the discussion "veered off into whales" is incorrect. The discussion was about the accuracy of certain paleontological methods, and that paper vetted those methods. That's not "veering", it's directly addressing the issue in question. That the paper was about whales rather than Neanderthals is of little consequence unless someone can give a reason for why that distinction matters.

We assumed, meaning that figure was plucked randomly out of the air, that 10% of the Neanderthal fossils found should indicate some kind of deformity if 10% of the HSS population had that incidence rate. The reason being that Neanderthals are closer to us genetically than chimps, there would probably not be a huge statistical difference. We assume that the total population for Neanderthals was 70,000 with only something like 400 fossils found so far. Is that enough to obtain any kind of idea of what would indicate variation or deformity or would you assume it was a whole other species if you did find something like that?

You appear to have misunderstood the thought experiment. The question isn't "do we have enough data to have an idea of what would indicate variation or deformity or would you assume it was a whole other species if you did find something like that?" The question is far more simply, "is it reasonable to expect to ever find a deformed fossil in the first place?" If it's not reasonable to expect to ever find a deformed fossil, and you find a fossil that is either a deformed Neanderthal or a new species, then it's more reasonable to assume it is a new species. Note that this doesn't take into account paleopathology at all, and instead assumes (incorrectly, but for the sake of argument) that we have no way to tell if bones are deformed are not.

My scenario might not be a relevant edge issue in paleontology but it would be in my modern day work world where you are trying to decide which populations and what disease incidence warrants allocation of funds.

We're not discussing your work. We're discussing paleontology.

I'm not saying what you are doing is wrong

How are you not? You're saying they don't have enough fossils for it to be a "representative sample". If that's not saying they're doing it wrong, then what is it? Either you're criticizing the field or you're not.

 

[Jodie]

The problem is that you haven't explained how large is "large enough", or how you make that determination. According to Dinwar most species have very little morphological variation, and so a huge sample size isn't needed. You've given us no reason to think this is wrong.

Once again an assumption, but it would depend on how big you think the population size was in comparison with the number of fossils found to determine how many would be enough.

As has been explained multiple times, that paper is directly related to the topic at hand. Saying that the discussion "veered off into whales" is incorrect. The discussion was about the accuracy of certain paleontological methods, and that paper vetted those methods. That's not "veering", it's directly addressing the issue in question. That the paper was about whales rather than Neanderthals is of little consequence unless someone can give a reason for why that distinction matters.

And they had a large number of whale fossils and whale DNA. This isn't always the case with humans. It is misleading because you're taking a set of circumstances or data that is more complete in the example of the whales than what you have for most hominid species. The conclusions are only as good as the data you have to work with.

You appear to have misunderstood the thought experiment. The question isn't "do we have enough data to have an idea of what would indicate variation or deformity or would you assume it was a whole other species if you did find something like that?" The question is far more simply, "is it reasonable to expect to ever find a deformed fossil in the first place?" If it's not reasonable to expect to ever find a deformed fossil, and you find a fossil that is either a deformed Neanderthal or a new species, then it's more reasonable to assume it is a new species. Note that this doesn't take into account paleopathology at all, and instead assumes (incorrectly, but for the sake of argument) that we have no way to tell if bones are deformed are not.

If you have a modern day skull with a sagittal keel is it a deformity? No, it's a variation of a modern skull found in a fraction of the HSS population but you also find this in Homo Erectus.

We're not discussing your work. We're discussing paleontology.

It was relevant to the reference Dinwar made in how statistics are used across fields, but by all means, if you want to undermine my comment be my guest.

How are you not? You're saying they don't have enough fossils for it to be a "representative sample". If that's not saying they're doing it wrong, then what is it? Either you're criticizing the field or you're not.

If you think the estimate of 70,000 Neanderthal as the total population that existed is correct then 400 fossils wouldn't be a large enough sample to know how much variation actually existed within the species. Although it's still tentative, the article I linked is discussing the new Homo Erectus skull and how it places all the subspecies into question. I'm assuming those subspecies were identified strictly through morphology since we have no Homo Erectus DNA of any kind for comparison. I'm criticisizing the field because it has it's limitations for the above stated reasons and I'm not incorrect in doing so.

 

[Dinwar]

We started to address that with the Neanderthals at the beginning of the thread but the thread veered off into whales. We assumed, meaning that figure was plucked randomly out of the air, that 10% of the Neanderthal fossils found should indicate some kind of deformity if 10% of the HSS population had that incidence rate. The reason being that Neanderthals are closer to us genetically than chimps, there would probably not be a huge statistical difference. We assume that the total population for Neanderthals was 70,000 with only something like 400 fossils found so far. Is that enough to obtain any kind of idea of what would indicate variation or deformity or would you assume it was a whole other species if you did find something like that?

If you knew half as much about statistics as you think you do, you would realize that this has nothing to do with the question I asked. I've already explained how the whale paper demonstrates that paleontological data and genetic data are identical where they overlap, meaning that your innane assertion that we must use genetic data not only is unsupported, but actually contradicts the data.

I have wasted far more time on you than your ideas--which beligerantly ignore entire fields of science--deserve. I'm done. http://palaeo-electronica.org/content/resources Here is a linke to Paleontologia Electronica, one of the paleo journals that is most likely to be sympathetic to your views. Please submit a paper outlining your views to this journal--the proper venue for these discussions. I will keep an eye on it, and will respond in kind.

As a bit of friendly advice, you may want to do some basic research on the fields you are attacking (and yes, saying that entire fields YOU KNOW NOTHING ABOUT and refuse to learn about can't do their jobs IS attacking, whether you want to admit it or not) before submitting a paper. You will find, if you fail to do that, that I'm actually one of the nice guys.

 

[Akri]

Once again an assumption, but it would depend on how big you think the population size was in comparison with the number of fossils found to determine how many would be enough.

First, what are you calling an assumption? I can't tell if you're referring to one of my statements as an assumption, or if you're saying that you could only determine the needed sample size via assumption.

Second, please walk me through this process. We have a population of estimated size X, how many fossils do we need? How does the rate of expected morphological variation factor into that?

And they had a large number of whale fossils and whale DNA. This isn't always the case with humans. It is misleading because you're taking a set of circumstances or data that is more complete in the example of the whales than what you have for most hominid species. The conclusions are only as good as the data you have to work with.

I think you missed the point of that paper. The point was that when we're able to compare the paleontology data to genetic data, the two agree with one another. Thus the paleontology methods are sound.

The methods used in the whale paper are the same methods that are used to study Neanderthals. If the methods work with whales then they work, period. Unless you can actually demonstrate that the methods only work with a certain sample size? I don't think you've actually demonstrated such a thing yet; you've merely asserted it.

If you have a modern day skull with a sagittal keel is it a deformity? No, it's a variation of a modern skull found in a fraction of the HSS population but you also find this in Homo Erectus.

I'm not sure what this has to do with me pointing out that you misunderstood the thought experiment.

It was relevant to the reference Dinwar made in how statistics are used across fields, but by all means, if you want to undermine my comment be my guest.

You wrote it in response to his quote explaining the process of speciation, and how people only think speciation events in hominids are important because we're hominids. Perhaps you meant to make your statements in response to something else?

If you think the estimate of 70,000 Neanderthal as the total population that existed is correct then 400 fossils wouldn't be a large enough sample to know how much variation actually existed within the species.

That's not what the thought experiment was asking, though. The questions was, if a certain variation definitely affected X% of the species (regardless of whether we're aware of it or not), what are the odds that we would ever actually see a fossil displaying this variation? That is a very different question than whether or not we have enough fossils to know how much variation existed.

Although it's still tentative, the article I linked is discussing the new Homo Erectus skull and how it places all the subspecies into question. I'm assuming those subspecies were identified strictly through morphology since we have no Homo Erectus DNA of any kind for comparison. I'm criticisizing the field because it has it's limitations for the above stated reasons and I'm not incorrect in doing so.

OK, so you are telling paleontologists that they're doing it wrong.

Thing is, you haven't actually demonstrated that their methods are problematic. You've said that the sample sizes they use are too small, but you haven't shown any actual evidence that this is a problem (at least, not that I recall). It isn't enough to say "the sample sizes are too small". You need to show actual evidence of the small sample sizes leading to incorrect conclusions.

Additionally, other data has been presented showing that 1) where paleontological data and genetic data overlap the two corroborate one another, and 2) an entire field of study revolved around identifying fossils that are actually deformed as opposed to being a different species. You need to take both of these concepts into account in your criticisms of paleontology, and thus far I don't think you've done so.

 

[Dinwar]

I think you missed the point of that paper. The point was that when we're able to compare the paleontology data to genetic data, the two agree with one another. Thus the paleontology methods are sound.

To be fair, the point of the paper was to define the nature of baleen evolution. That said, the figures I pointed out do serve as a test to determine the efficacy of morphological data in determining evolutionary history.

Unless you can actually demonstrate that the methods only work with a certain sample size?

I'm going to go further: given the nature of the mutations they were examining, sample size is irrelevant. It doesn't matter if you had one or five thousand samples--because these mutations are necessary aspects of those organisms, EVERY organism in that taxa will have those mutations (except extremely rare genetic freaks, who likely died in vitero). It's akin to the broken Vitamin C gene in humans: since it's universal, a single sample is sufficient.

Think of it this way: If you flip a 2008 D penny, the odds of the penny being made with a copper coating are 1. The odds of it coming up heads or tails is irrelevant, and we don't need to flip the coin ten thousand times to figure that out. Similarly, the odds of morphological diversity that has no bearing on the traits definitive of a species have no bearing on the necessary sample size to determine the species' morphology. Certain traits are universailly present in certain species. Sometimes we err in our assessment of universality, but that's why we continue to look for fossils of species we've already described. None of that makes paleontology subjective--you cannot be so empirical that you become subjective.

 

[Akri]

To be fair, the point of the paper was to define the nature of baleen evolution. That said, the figures I pointed out do serve as a test to determine the efficacy of morphological data in determining evolutionary history.

By "the point of the paper" I really meant "the point of mentioning that paper in this conversation."

I'm going to go further: given the nature of the mutations they were examining, sample size is irrelevant. It doesn't matter if you had one or five thousand samples--because these mutations are necessary aspects of those organisms, EVERY organism in that taxa will have those mutations (except extremely rare genetic freaks, who likely died in vitero). It's akin to the broken Vitamin C gene in humans: since it's universal, a single sample is sufficient.

Think of it this way: If you flip a 2008 D penny, the odds of the penny being made with a copper coating are 1. The odds of it coming up heads or tails is irrelevant, and we don't need to flip the coin ten thousand times to figure that out. Similarly, the odds of morphological diversity that has no bearing on the traits definitive of a species have no bearing on the necessary sample size to determine the species' morphology. Certain traits are universailly present in certain species. Sometimes we err in our assessment of universality, but that's why we continue to look for fossils of species we've already described. None of that makes paleontology subjective--you cannot be so empirical that you become subjective.

If you do err in your assessment of universality, how would you determine that? If you've identified feature X as being universal in species Y, but then you find what appears to be a member of species Y without feature X, how do you determine if it's a new species vs. you being wrong about feature X applying to all members of species Y?

 

[Dinwar]

If you do err in your assessment of universality, how would you determine that?

We look for more specimens and adjust our understanding based on what we've found. For example, "Decapoda" means literally "ten legs", and that's often taken as one of the diagnostic traits of the taxa. However, there are crabs that have secondarily lost their hind-mostlimbs,resulting in 8-legged decapods.

As for species, it's trickier. It depends on how vital that trait is. If it's something that's often diagnosticin the taxa--say, tooth morphology in mammals--it may be sufficient grounds for identifying a new species. If it's something that's highly variable within the group--say, digit number in the first tetrapods (which had different numbersof digits on different limbs of the same organism in some cases)--it's likely to be called a subspecies or mere variation within the species. This does not mean that researchers simply decide at random, or subjectively; however, since bauplans are different between phyla and often fairly variable within them, it really does require a certain amount of expertise to know what traits constitute sufficient grounds to call something a new species. You simply need to be immersed in the morphology to understand what traits are important and what traits aren't. Once you have that deep understanding, the question has objective answers.

 

[Jodie]

Well, I find the genetic part more reliable. You don't have that most of the time in hominid paleontology. I chose the Neanderthals as an example because we do have their DNA. Now that we have Neanderthal DNA we have a better idea of how they relate to us, although there is still conflicting theories floating around about what that DNA actually indicates.

 

[Akri]

Well, I find the genetic part more reliable.

Based on what data?

 

[Dinwar]

Based on what data?

There can't be any. I've demonstrated that the genetic data and the morphological data produce the exact same results where they overlap (the only rational way to evaluate them). Furthermore, genetic data is necessarily limited to only the most recent species, and a few Holocene to Pleistocene bones found in extraordinary conditions.

The issue with using DNA data to reconstruct phylogenies is obvious: DNA data misses the overwhelming majority of species. By some counts, the number of species we can get DNA data from is statistically insignificant. Using only DNA we can NEVER understand evolutionary history. There are far, far, FAR too many holes. And since morphology yields the same results, we have no reason to discard that dataset--except for the uninformed opinion of one person.

 

[Jodie]

Dinwar, I didn't say discard it, it's all you have. I just think relying on morphology alone when you don't have a high enough percentage in the fossil record to represent the population isn't going to give you an accurate idea of who and where we came from if DNA is out of the question. DNA technology improves exponentially every year. Who is to say once we unravel it all that it won't be more informative piecing the story together from living taxa, much like you do already with morphology, but less subjective IMO.