taxonomy, DNA and Scilla

[Jim McKenney]

I read this today in an atricle about splitting up the genus Scilla to make more sense of it.

Finally, what is left of Scilla includes species such as S. bifolia as well
as the now defunct genus Chionodoxa (which itself forms 2 unrelated groups).

It does make sense of the whole fertile intergeneric hybrid thing with Scilla and Chionodoxa. I'm not normally a splitter by the way.
See http://lists.ibiblio.org/pipermail/pbs/2003-December/016258.html

Simon, some of us take it a step further and say that the ability of nominal species to "hybridize" is proof that the entities in question actually make up one morphologically variable species.

Traditional taxonomy was based on morphology. The traditional view was, more or less, if two things looked alike, they were members of the same species.

But that point of view got it backwards: if two organisms look alike, the likely explanation is that they share the same gene pool. In other words, they are not members of the same species because they look alike, it's rather that they look alike because they are members of the same species and share the same gene pool. (There can be other explanations, too.)

And of course, the members of some widely distributed species which form Rassenkreise or clines sometimes do not resemble the other members of their species from distant parts of the range of the species.  

Natural "hybrids" are often best regarded as an indication that the purported parents are conspecific.

Now back to the fun stuff. Here's a lovely little plant which appeared in my lawn this week. It looks like a super Scilla bifolia, but I suspect that it's had some help from erstwhile Chionodoxa sardensis. What do the rest of you think?

[Sinchets]

I suppose the problem has always been that assuming 2 things that look similar are similar. Maybe this is why the 'New World' and the Southern hemisphere are full of wrens and blackbirds. I.e. find a species of bird no-one from Europe has seen before:- Is it small? Ah must be a wren. Is it black? Ah must be a blackbird. I guess that's how we end up with Scilla species on several continents ;- Is it blue (ish)- Must be Scilla and maybe helps explain a pan-global Viola genus.
If more DNA research continues to find Amaryllidaceae really has evolved from Alliaceae we will have to accept that we cannot judge by looks alone.

[gote]

Simon, some of us take it a step further and say that the ability of nominal species to "hybridize" is proof that the entities in question actually make up one morphologically variable species.

Jim,
That would make big chunks of orchidaceae into one species. A surprising number of orchids are able to interbreed.
Göte

[Jim McKenney]

Simon, some of us take it a step further and say that the ability of nominal species to "hybridize" is proof that the entities in question actually make up one morphologically variable species.

Jim,
That would make big chunks of orchidaceae into one species. A surprising number of orchids are able to interbreed.
Göte

Yes, Göte, that's precisely the point I made on another forum in the past when the topic of the byzantine tangle of intergeneric hybrids of orchids arose: that some of us would cite the ability of these plants to produce viable, sexually reproducing progeny as evidence that the parental "genera" are in fact one morphologically variable species.

[Gerry Webster]

Simon, some of us take it a step further and say that the ability of nominal species to "hybridize" is proof that the entities in question actually make up one morphologically variable species.

Traditional taxonomy was based on morphology. The traditional view was, more or less, if two things looked alike, they were members of the same species.

But that point of view got it backwards: if two organisms look alike, the likely explanation is that they share the same gene pool. In other words, they are not members of the same species because they look alike, it's rather that they look alike because they are members of the same species and share the same gene pool.

Jim - chicken & egg. In the case of animals, two organisms share, or contribute to, the same gene pool either because they 'look alike' (to each other) or are in some way mutually recognizable. In the case of plants, they are 'mutually recognizable' in the sense of being morphologically or physiologically compatible. The so-called 'Biological' concept of species has as many problems as the older morphological concept.

[Jim McKenney]

Simon, some of us take it a step further and say that the ability of nominal species to "hybridize" is proof that the entities in question actually make up one morphologically variable species.

Traditional taxonomy was based on morphology. The traditional view was, more or less, if two things looked alike, they were members of the same species.

But that point of view got it backwards: if two organisms look alike, the likely explanation is that they share the same gene pool. In other words, they are not members of the same species because they look alike, it's rather that they look alike because they are members of the same species and share the same gene pool.

Jim - chicken & egg. In the case of animals, two organisms share, or contribute to, the same gene pool either because they 'look alike' (to each other) or are in some way mutually recognizable. In the case of plants, they are 'mutually recognizable' in the sense of being morphologically or physiologically compatible. The so-called 'Biological' concept of species has as many problems as the older morphological concept.

Gerry, when you say "The so-called 'Biological' concept of species has as many problems as the older morphological concept" , isn't it true that the "problems" are largely of a utilitarian nature? What I mean is that people in general expect species concept to be useful in some way, not simply a reflection of what is happening in nature. 

And as for "chicken & egg": you must not be a zoölogist. Zoölogy answered this one long ago. The egg definitely came first. Chickens lay what is known as a cleidoic egg. The cleidoic egg evolved in the reptilian ancestors of chickens - it was around long before there were chickens. 

[Gerry Webster]

Gerry, when you say "The so-called 'Biological' concept of species has as many problems as the older morphological concept" , isn't it true that the "problems" are largely of a utilitarian nature? What I mean is that people in general expect species concept to be useful in some way, not simply a reflection of what is happening in nature. 

And as for "chicken & egg": you must not be a zoölogist. Zoölogy answered this one long ago. The egg definitely came first. Chickens lay what is known as a cleidoic egg. The cleidoic egg evolved in the reptilian ancestors of chickens - it was around long before there were chickens. 

Jim - Yes, there are problems in use by 'people in general' but I was referring to the scientific /philosophical problems pertaining to what species really are in nature. Not really a topic for a short post. I can PM you some references if you are interested but you will need access to a library which has a good range of journals, preferably a university library.

Well, my first degree was in zoology (of which I remember virtually nothing - it was a a very long time ago) but I was speaking metaphorically. The 'gene pool' is a dynamic entity which needs to be maintained & reproduced.

[Sinchets]

Just because. Do you ever wonder what would happen one day if a Cattleya orchid (asa humanologist)was ever asked to try and attribute a species name to humans? Would it collect a fair selection of representative forms of human? Would it put all of these in one species, even though they may have red, brown, black or blond hair- how would it differentiate between differences in skin colour or height, body frame and weight? Would it maybe take a step back and check to see if humans were able to 'hybridise', and then wonder why the offspring generally didn't look much like their parents? The answer I guess is in the DNA somewhere.

[Gerry Webster]

The answer I guess is in the DNA somewhere.

Many, many years ago on the BBC Radio 4 programme 'Any Questions' a regular participant was a farmer named A.G. Street. No matter what the question, his response was always - "The answer lies in the soil". He became famous. 

[Jim McKenney]

The answer I guess is in the DNA somewhere.

Maybe, but maybe not.

Each generation of scientists has seized on new advances in  technology and applied these to the question of species. These new technologies typically provide insight, usually the sort of insight which allows us to say "A and B are not the same species". But no technology that I'm aware of allows us to say, purely on the basis of morphology, that A and B are the same species. Conspecificity is a function, a relationship, not a structure. It is tempting to infer function from morphology, but it's very difficult if not impossible to prove it purely from morphology.

I'm willing to bet that when the dust settles, current DNA studies will prove to be another example of this. DNA studies after all are a much more finely granular morphology - really  a sort of meta-morphology. Is there really any reason to think that DNA studies, to the extent that they are ultra-refined morphological studies,   will provide the answer to what is essentially not a question of morphology?

My apologies to all who have read this far and are wondering what any of this has to do with Scilla.

[Gerry Webster]

The answer I guess is in the DNA somewhere.

Maybe, but maybe not.

Each generation of scientists has seized on new advances in  technology and applied these to the question of species. These new technologies typically provide insight, usually the sort of insight which allows us to say "A and B are not the same species". But no technology that I'm aware of allows us to say, purely on the basis of morphology, that A and B are the same species. Conspecificity is a function, a relationship, not a structure. It is tempting to infer function from morphology, but it's very difficult if not impossible to prove it purely from morphology.

I'm willing to bet that when the dust settles, current DNA studies will prove to be another example of this. DNA studies after all are a much more finely granular morphology - really  a sort of meta-morphology. Is there really any reason to think that DNA studies, to the extent that they are ultra-refined morphological studies,   will provide the answer to what is essentially not a question of morphology?

My apologies to all who have read this far and are wondering what any of this has to do with Scilla.

Jim - I wonder whether you are  between Scylla & Charybdis? Yes, I agree that DNA analysis is a technologically advanced version of traditional morphological analysis.  Its limitations remain to be seen, though classical genetics suggests there will be some problems about relating it to gross morphology.

I don't  really understand what you mean by "conspecificity is a function", unless you are referring to the 'biological'  concept & the central importance there of the ability of two individuals to interbreed. The traditional, morphological concept of species - where a species is a class - is empirically untenable but there is nothing incoherent or logically flawed  about it. To speak of two individuals being morphologically similar  is to speak of  a relationship. Similarity is a relationship & classes are constructed on the basis of similarity or sameness in particular respects.

[Jim McKenney]

The answer I guess is in the DNA somewhere.

Maybe, but maybe not.

Each generation of scientists has seized on new advances in  technology and applied these to the question of species. These new technologies typically provide insight, usually the sort of insight which allows us to say "A and B are not the same species". But no technology that I'm aware of allows us to say, purely on the basis of morphology, that A and B are the same species. Conspecificity is a function, a relationship, not a structure. It is tempting to infer function from morphology, but it's very difficult if not impossible to prove it purely from morphology.

I'm willing to bet that when the dust settles, current DNA studies will prove to be another example of this. DNA studies after all are a much more finely granular morphology - really  a sort of meta-morphology. Is there really any reason to think that DNA studies, to the extent that they are ultra-refined morphological studies,   will provide the answer to what is essentially not a question of morphology?

My apologies to all who have read this far and are wondering what any of this has to do with Scilla.

Jim - I wonder whether you are  between Scylla & Charybdis? Yes, I agree that DNA analysis is a technologically advanced version of traditional morphological analysis.  Its limitations remain to be seen, though classical genetics suggests there will be some problems about relating it to gross morphology.

I don't  really understand what you mean by "conspecificity is a function", unless you are referring to the 'biological'  concept & the central importance there of the ability of two individuals to interbreed. The traditional, morphological concept of species - where a species is a class - is empirically untenable but there is nothing incoherent or logically flawed  about it. To speak of two individuals being morphologically similar  is to speak of  a relationship. Similarity is a relationship & classes are constructed on the basis of similarity or sameness in particular respects.

Yes, I had the biological concept in mind when I wrote that, and it was not my intent to use the terms function, similarity and relationship in the way they are used in formal logical argumentation.

Since we both agree that the the morphological concept of species is empirically untenable, I see no purpose in this argument in celebrating the logical validity of its structure or application. But please don't misunderstand me: in other contexts an understanding of those things can be important, if only because they help us understand the work of those who have worked from the viewpoint of the traditional morphological concept of species.

I think there is an interesting parallel between the discussion we are conducting and one concerning a knowledge of classical Latin and Greek in taxonomy. Such knowledge seems to becoming rarer and rarer among practicing horticulturists and botanists. It seems to me that even among those who regard such knowledge as irrelevant and deservedly obsolescent, some level of knowledge is necessary, if only because it will allow us to understand why things were done the way they were in the old days.

I'm all for preserving an understanding of why people in the past did things the way they did.

[Gerry Webster]

Since we both agree that the the morphological concept of species is empirically untenable, I see no purpose in this argument in celebrating the logical validity of its structure or application. But please don't misunderstand me: in other contexts an understanding of those things can be important, if only because they help us understand the work of those who have worked from the viewpoint of the traditional morphological concept of species.

Jim - I promise this is my last post on the subject. Lovers of Scilla must be fed up.

I claimed that the traditional morphological concept of species - based on gross morphology - is empirically untenable. This does not rule out a non-traditional morphological concept. A few biologists are attempting to develop such a concept  based on developmental mechanisms. It remains to be seen whether this is possible. Since the 'biological' concept has a multitude of problems all it's own, some alternative concept is required if we are to retain a notion of species as real entities.

[Jim McKenney]

Since the 'biological' concept has a multitude of problems all it's own, some alternative concept is required if we are to retain a notion of species as real entities.

It's another example of the old philosophical conundrum: as a concept approaches certainty, it loses meaning in the real world.

I'll bow out on that, and in doing so I hope to dodge both Scylla and Charybdis, although I wonder if either Scylla or Charybdis has an indigenous species of Scilla of the Oncostema sort.   

[jomowi]

Nothing is straightforward, Ramonda x Jancaea and the Ramoda x Haberlea hybrids both differ in chromosome numbers, hardly evidence of conspecificity.  

What about convergent and divergent evolution, the former when two clearly distinct species evolve to become look alikes?

I always understood that to be a species there must be some barrier to breeding.  Not even that is straightforward when even separation by large distances and long evolutionary time does not separate species that are clearly different.