what does something need to be a "New" Species?

Crash probably provided an good enough explanation in answer to DC's question.I agree with WK, tho'. Reproductive isolation - whether pre- or post-zygotic - is the key criteria to determining the existence of separate species (in sexually reproducing organisms, natch'). Geographic isolation can be a condition that allows reproductive isolation to arise in separated populations, but doesn't in and of itself imply speciation. In that case, you would have to test the species by seeing what happens if the geographic barrier is removed: if they reproduce, they're the same species. If they don't (or if F1, F2, Fn hybrids are unviable), then they're different species.It all comes down to gene flow: if two or more populations can have "significant" gene flow between them, then they're the same species. If the gene flow drops beyond a certain, undefined point, they're different species. This is what happens with ring species, for instance. There is significant gene flow between adjacent populations, but by the time the two ends of the chain get back together, reproductive barriers have arisen and they aren't considered the same species. It gets pretty arbitrary (where do you draw the imaginary line?). Which of course is why the creationists' essentialism is bogus.

Species are best defined as 'recognisably seperate animal types'. The concept of species was been used for a long before ANY attempt was made to define it, and the definition I've just given most closely approximates this common law usage.Personally I think a Jack Russell and a Rottweiler SHOULD be classified as different species, they are not simply by merit of their artificial status.Reproductive isolation is not only very hard to use 'in the field' (as we cannot exhaustively mate one species with another to test it), but fails to distinguish between groups generally recognised as being different species (wolf and dogs being a common example - many, many species of butterflies being a less common one). Further it is quite useless when it comes to classifying extinct animal types.

quote:

---

If they don't (or if F1, F2, Fn hybrids are unviable), then they're different species.

---

Hi Q,
I am failing to see how you are defining reproductive isolation here. F1, F2,Fn implies that the two "species" are not reproductively isolated and can actually interbreed and in the case of F2 to Fn..the hybrids can breed several generations...I prefer what you said about levels of gene flow among populations as a criteria..if two adjoining populations that can produce offspring do so seldomly..I would say they are separate species whether it be behavior or whatever that is inhibiting the transfer of alleles between the two groups.An example of this would be L.cyclotis and L. africana. There are slight morphological differences between the two elephant species..where they come in contact they sometimes produce hybrid offspring...but if you sample them with multiple genetic markers from microsats to specific gene sequences, they have almost as many fixed differences between them as Asian elephants have from both.

Oh goody, a disagreement that isn't based on creationist propaganda and misinformation. MrJ: The morphological species concept you espouse has its uses. It is used fairly widely, as I understand it, to "type" fossils, for instance (where the biological species concept is unworkable). However, in living systems it tends to render more false positives (or false negatives, for that matter) than the biological species concept. Consider convergent evolution, for instance, which was one of the key reasons for casting about for a different way of classification. Even your own example, butterflies, shows how unworkable the morphological concept is. Consider all of the tasty Nymphalidae (like, fr'instance, Eresia spp) that are near-perfect mimics of toxic Heliconidae (Eresia looks exactly like Heliconius erato). Having experienced this mimicry first hand (my exchanges with my partner on field trips often ran something like, "And just how the HELL did you know that was a different species, from 5 meters away? Even in the net they all look alike to me!") Worse, they share the same habitat and often the same congregation. Even the experts occasionally have to rely on genetic typing to differentiate butterflies.By the same token, I disagree with you that reproductive isolation is overly hard to test in the field (except in the case of wide geographic isolation, in which case you have to rely on genetics). All it takes is observation: determining whether there is overlap in ranges between two populations, and if so, is there a hybrid zone or not between them? If they overlap and there's no gene flow, then they're distinct species - no matter how much they might resemble one another. If there's a hybrid zone, then how viable are the hybrids in comparison to parent populations? If they are declining inspite of the potential for dispersal from the two parent populations, then (ceteris paribus) the parent populations may be in the process of speciation.Finally, if two populations ARE geographically isolated such that gene flow is physically impossible, then genetics may give you the answer. IOW, in this case if the karyotypes are vastly different, you might have a case for splitting the two populations into subspecies or species. OTOH, you'll get a major argument from the "lumpers" when you do... As to your comment concerning dogs - I agree. The principal reason that major dog breeds aren't considered separate species is precisely because it is artificial. One of the main elements in the biological species concept is that the reproductive isolation applies "in the wild". If you had wild populations of Jack Russells and Rotties adjacent to each other, and they still didn't breed, then they WOULD be different species. Besides, dogs are just wolves with pretensions...

I know, I know. The only reason I even include hybrid inviability past F1 in the criteria is because someone threw it back in my face once when I asserted otherwise. It was something along the lines of, "You're ignoring the fact that in this (don't ask) Drosophila experiment the F1 hybrids were fully fertile, and the hybrid inviability didn't show up until F2." Yada, yada, yada. So to avoid quibbles, I add the Fn bit. I personally STILL consider it waaaay down on the list of possible post-zygotic barriers/criteria. And I've never heard of a case study where this occurred in the wild. If everyone here will agree that hybrid inviability in F1 in natural populations should be the criteria, I'm quite happy to drop the rest. Until the next smartarse comes up with that fly experiment. As to the populations bit - well, you know me. Populations R Us. So I completely agree with you here: This comes back to where to draw the line and say gene flow is no longer "significant".

I'll stop you there. I don't espouse a morphological species concept, I espouse a concept based on recognisable differences - whether purely morphological, behavioural, or genetic. A demonstration of reproductive incompatability would be an entirely valid seperation under the scheme I espose.I also espouse a concept defined by common agreement rather than rigid definition.While the absence of hybrids does provide evidence of reproductively seperate species, it doesn't give conclusive evidence. Two species could simply not recognise the other as potential mate, even though they would successfully mate if they were ever paired.Finally, I would like to know what you consider the correct approach when the biological approach yields results that do not match with accepted species definitions?

All you seem to be saying is that species are different from each other in a variety of non-specific ways. This seems to simply be a case of your species concept being a superclass of the various already extant species concepts but with nothing else. You are just saying 'we will consider these seperate species if they satisfy the criteria of the BSC or the MSC or the ESC etc...', which doesn't really leave us any better of than before.If two populations fail to recognise each other as potential mates and therefore don't mate then where do the hybrids come from. It has already been pointed out that the criteria are based more on the behaviour of the animal in the wild than what the limits are that you can reach experimentally in the lab.The BSC approach obviously yields results that match the BSC definition, which accepted defnition are you thinking of?

Many species of canines can interbreed (wolf, arctic wolf, dog, coyote - I think, african hunting dog). Some species of cat can interbreed. An article in the New Scientist, I think it was early last year, discussed hybridisation among Butterflies in the rainforest, specifically that many different species could interbreed giving viable hybrids. Although I'm afraid I neither have an exact reference to give you, nor easy access to back issues to search through for one.The concept of species predates the BSC defintion, the MSC definition and the ESC definition. I consider that any attempt at a definition that fails to give the distinction found in this earlier concept has failed to give a proper defintion of this concept.My opinion is that we will never be able to satisfactorily define species because the concept is a human ordering not actually present in the real world.

Hi Mr J: Okay, I stand corrected. You have to admit, however, that the way you phrased your opening remarks: espcially the bit about "recognisably separate animal types" and the traditional, "common law" nature of the definition you're using is a close match for what is typically described as the morphological concept. Cronquist's definition of the morphological/phenetic species concept is "... the smallest groups that are consistently and persistently distinct and distinguishable by ordinary means." (from here). It's an updated version of the Linnean system, which is where your note of "common law" took me. Sorry for the confusion.It does sound, however, that you are using Cronquist's phenetics as a basis. Is that a correct assumption? Perhaps you could expand a bit, so I don't keep flailing around in the dark? Thanks. Hey, me too! What d'ya know. BSC is a pretty "open" definition, I'd have thought. Reproduction is about the sole thread running through all sexually reproducing organisms. I mean, if you're going to adopt a definition, shouldn't it be as generally applicable as possible? There are substantial problems with the BSC, not the least of which is its limited applicability to plants and its nearly universal inapplicability to single-cell, hermaphrodite, or clonal organisms. I'm not sure the entire idea of "species" itself is a valid one. OTOH, you do have to draw the line somewhere. How would your concept work in practice? I'm not sure I'm following you here. If two populations in contact where there is potential gene flow do not interbreed through behavioral differences ("not recognizing each other as potential mates"), then by definition they are reproductively isolated, and hence distinct species. When you say "if they were ever paired", are you implying a forced mating? If so, you've removed their success from BSC consideration, as they are artificial hybrids. One example is the range overlap between Bengal tigers (Panthera tigris) and the Asiatic lion (Panthera leo persica) in India (the Gir Forest, if you're interested). There are NO known natural hybrids. However, forced mating (actually, I think it was artificial) have bred viable hybrids. Under the BSC, they are considered distinct species. This is, however, certainly a case where two populations don't recognize each other as potential mates in the wild, and yet can have offspring.Perhaps that's not what you meant. If not, please clarify. Again, I'm not sure I'm following what you're asking. There are a number of examples where, after study, one species was taxonomically split into two (happens a lot with sympatry, which can be difficult to detect without study - a number of phytophagous beetle reclassifications come to mind). There are also examples where two populations, thought to be separate species, were combined when it was found that they weren't really-and-truly reproductively isolated. Specifics escape me at the moment. Or isn't that what you're asking?{edited to fix weird formatting mistakes}[This message has been edited by Quetzal, 08-13-2003]

Yes, my original posting was rather vague. My apologies. I'd say not, I didn't recognise the name Cronquist, although I had read that piece you posted a link to. Essentially I think reproductive, morphological, behavioural and, if available, evolutionary data are all valid grounds for differentiating a species. See also my remarks in post 23 above.Mostly I feel that BSC fails as definition of species both on some of the grounds you mentioned (inability to define species of asexual animals, general failure with plants) and it's tendency to group animals which would otherwise be considered different species (dogs/wolves). And that a better definition is therfore needed. A pure morphological definition fails in cases of convergent evolution and mimicry as you have stated. A pure evolutionary concept is too difficult to prove, demonstrate or apply in the field. I feel the best approach is to use an intelligently applied combination of all three, as well, possibly as other lines of evidence. I think I have misunderstood them. I thought BSC was 'can't interbreed' not 'don't interbreed'. However I think that in itself leads to further problems. For example in many island chains one can locate isolated populations of what would generally be considered the same species, but cannot be under your definition. Also there is the problem of at what level of gene flow do you consider it isolated. It is much easier to show hybrids are rare - unobserved during the (short) period of study - than never happen. Are dogs are wolves the same species? Are coyotes and wolves the same species? Are african hunting dogs and wolves the same species?Oh, new point. If we have a group A, that interbreeds with group B, so we get a new species AB. But group A does not interbreed with group C, while group C does interbreed with group B are A & C the same species?

Mr Jack, Quetzal and I have both directly addressed this problem already, that while geographic isolation enforces a type of reproductive isolation it is not the same as reproductive isolation if the two species are in the same area and will not, or cannot, breed.I'd say it is 'can't interbreed' as long as the situation you are looking at is the naturally occurring one in the population but not 'can't interbreed unless you monkey around experimentally'.AS to your 'new point' it looks like the usual ring species example, although your definition is somewhat murky, If a and b can interbreed then they will not produce a new ab species as they are already subspecies of the same species. Ring species are very interesting and as has been mentioned before, constitute a good example of incipient speciation. Or did you set up your question right? Was species ab supposed to be the same as species c?

I'll try and clarify then:Wolves and dogs can, and do, interbreed. Dogs and African Hunting Dogs can, and do, interbreed. Wolves, and African Hunting Dogs do not. These three are traditionally considered diferent species. I used the term 'group' to try and avoid getting a murky confusion between traditionally regarded as a species, and BSC-defined as a species. I used the term 'new species' to mean a different species as defined by BSC as compared to traditional classification.Now, BSC considers wolfs and dogs the same species. Yes? And therefore wolves, dogs and african hunting dogs the same species? Yes? So wolves and african hunting dogs WERE different species, but aren't NOW because we produced dogs from wolves? Do you not see that as a problem with the definition? I don't see how this works. If I get two species and breed them then that doesn't make them the same species, unless they don't meet in the wild? How can that work? It'll vary with time, according to the areas inhabited by the species at that time.

As I already suggested the wolf/dog/hunting dog example sounds just like the many cases of ring species where intermediate populations can interbreed with either extreme but the extremes are not themselves interfertile.Lots of things that were considered different species by different criteria have subsequently been lumped together and some have been split apart, that is bound to happen if you use a completely different definition for a concept that has been around for a long time. I'm not sure quite where the problem lies but I think you are misunderstanding me. What I am saying is that two interfertile populations which are solely reproductively isolated by geographical isolation but would mate if in the same location should not be considered seperate species but that species seperated by other mechanisms, such as changes in mating behaviour, which will not mate despite being in the same location should be considered different species even if they are capable of interbreeding through experimental intervention.The best way to avoid murky confusion would be to make it clear which concept of species you are using, I don't know that saying group make sit any clearer.

Wounded King writes:

quote:

---

What I am saying is that two interfertile populations which are solely reproductively isolated by geographical isolation but would mate if in the same location should not be considered seperate species but that species seperated by other mechanisms, such as changes in mating behaviour, which will not mate despite being in the same location should be considered different species even if they are capable of interbreeding through experimental intervention.

---

How's this for an example:St. Bernards that live in the United States should not be considered a different species from St. Bernards that live in the United Kingdom. There's an ocean that separates them and the population of St. Bernards in the US could easily interbreed with the population in the UK if they were to be put in proximity.However, if we were to eliminate all dogs except for pure-bred Chihuahuas and pure-bred Great Danes, a case could be made that these are separate species because even if they were in the same geographic locale, the two would not be able to interbreed due to physical constrants on the sexual act. You could artificially inseminate the Great Dane with Chihuahua sperm and come up with viable offspring (or so I'm asserting for the sake of this example), but the two will never do it on their own.[Please, don't try to come up with ways around this analogy like the Great Dane lying down or early puberty or some such. Just take it for the general image it is trying to convey.]Is that what you're getting at? That's the way I took it.------------------
Rrhain
WWJD? JWRTFM!

And I see that as a problem. As I see it, we have a concept of a species, and are now trying to concoct a definition that encapsulates that existing concept. Sure you can have new definition that doesn't match up, but to me that means your definition is wrong. This, as I see it, is the fundemental flaw with the BSC concept.You didn't answer my point about wolves and hunting dogs being diferent species before dogs, and the same species now. Do you not see that as a problem with the BSC definition? I do not see how this can be determined without experimental intervention. Yes you say you need would not consider them the same species if they could only interbreed with experimental intervention?