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

That is exactly the situation I was talking about Rrhain, but put in a much more digestible format. In fact I'm fairly sure we have had almost this exact same scenario fleshed out once or twice before in the old speciation thread. We've definitely encountered the St. Bernard's and chihuahuas in compromising positions concept before.If you haven't seen the old thread before Mr. Jack you might look at it here.[This message has been edited by Wounded King, 08-13-2003]

I'm not trying to sabotage the topic by any means, since I think everyone has made valid points so far. I simply want to point out that the definition of a species is exactly the sort of clear demarcation that Darwin's theory denies.The difference being one of extent and not essence is the hallmark of Darwinism. We only talk about species as a convenient shorthand that has to stand in for the complexity of biology. The difficulty of knowing where to draw such lines is what killed off the notion of created kinds, and with it species essentialism.I think there should be lengthy debate on this issue, simply because from a Darwinian perspective there's no easy answer. Similar debates rage all the time about the status of a dialect versus a distinct language, and the empirical investigation in this instance is similarly based on measuring degree of divergence. The magic point where two related organisms are said to belong to different species is a completely arbitrary one.------------------
En la tierra de ciegos, el tuerto es el Rey.

Hi again, Mr. J: I don't really disagree with this. My question, then, is exactly how this would work in practice? You could use a continuation of the Canid discussion (which I carry on below) as an example. The question would be something like: Under your conceptual framework, how would one determine the existence of a species within the New World representatives of family Canidae? What should the actual species be? Two things here: BSC says "don't interbreed in the wild". As for island biogeography, it doesn't really pose that much of a problem for the BSC. There are a couple of things that happen when an island is colonized from a source population (either on another island or on a mainland), depending on dispersal ability of the particular organism and some other factors. If dispersal ability is high, there may be no speciation on the island due to uninterrupted gene flow with the source population. OTOH, if gene flow is low or non-existent (and the colonizing population is minimally viable AND there are available niches - whether through vacancy or competitive exclusion), then a phenomenon called ecological release may occur resulting in rapid diversification and adaptive radiation: the so-called founder effect. In this case speciation may occur multiple times (depending on the carrying capacity of the island). Evidence for declaring the critters on the island a different species from its parent would derive from the low dispersal AND behavioral or phenotypical variance from the source. The presumption of speciation in this case is based on the fact of geographic isolation + phenotypical variance. Confirming the presumption WOULD require some kind of hybridization test. However, the working hypothesis under the BSC is that the longer a population has been isolated from gene flow (like on an island), the greater the probability that it has diverged from its parent population to the point where it could (not necessarily would) be proclaimed a unique species. I hope you'll note from this that the BSC is not precisely a rigid definition of species - it's a working model. Ah yes, the Canidae. It's a family that, especially in genus Canis truly defies any rational taxonomy. It is truly one example where the BSC seems to break down. I'm setting the African wild dog aside for the moment. Not that I have any particular prejudice against Lacaon pictis, but rather I simply don't have any decent references that discuss possible hybridization with domestic dogs. I'm also ignoring for the sake of this discussion the smaller canids like the kit fox (Vulpes aelox), arctic fox (Alopex lagopus), and grey fox (Urocyon cinereoargenteus), because there are no known fox-dog hybrids, and after all, this is all about dogs...Taking the four commonly recognized North American (hey, I'm yankee-centric in my reference material, so sue me - it's what I'm familiar with) species of the genus Canis: the grey wolf (C. lupus), the red wolf (C. rufus), the coyote (C. latrans) and the domestic dog (C. familiaris - yeah, I'm aware this is disputed, bear with me), can we make a decision as to whether, under the BSC, these represent valid species?1. Looking at familiaris and lupus first. A quick perusal of google shows a number of websites and references indicating that domestic dogs and wolves can hybrize and produce viable F2 hybrids (although apparently they ultimately tend to backcross to more and more wolf-like). Therefore, under the BSC, they would be considered the same species. Moreover, since ultimately the hybrids tend toward wolf over the generations, the domestic dog should be reclassified as C. lupus. Can we confirm this? Well, it turns out that restriction fragment analysis of 6 dog breeds and 22 wild wolf populations show that dogs differ from wolves in less than 0.2% of their genomes. Cool! Dogs really ARE wolves. One down.2. C. rufus poses a few problems, not least of which is the fact that it became extinct in the wild in 1975. However, captive breeding programs showed that the species would breed true - the offspring were "typical" red wolves - halfway in size between coyote and wolf. Under the BSC therefore, it should be classified as a distinct species. It was something of a shock to the scientists studying the preservation of this species when they discovered that every single specimen was either a wolf or coyote hybrid and contained genotypes that would - if strictly adhered to - reclassify every known specimen (both from the captive breeding program of the last southern red wolves from Florida to the blood samples taken in 1975 from other wild populations to the pelts originally taken in 1901). All showed either wolf or coyote genotypes. And yet, this mixed breed bred true - red wolves kept turning out red wolves, whether based on coyote genotype or wolf genotype. It is, however, a genuine species under the BSC - it breeds true. It is a chimera in that it is an example of a new species created from interspecific hybridization.3. C. lupus vs C. latrans. Are they distinct species? Under rare circumstances, the two species CAN hybridize and both can hybridize with dogs. Under the BSC it would be a judgement call as to whether or not they should be considered distinct. However, until the arrival of European agriculture (and subsequent elimination or partial extinction of the grey wolf from major areas of its former range), coyotes and wolves were in direct competition - not unusual with closely related species - with the wolves exercising exclusion in the zones where they predominated. Coyotes were limited primarily to the south-eastern US. Today the opposite occurs - coyotes, supreme opportunists, occupy a range that extends across North America. However, since they can hybridize, should they be classified as the same species? In this case we do have to turn to genetics to confirm our plan to eliminate one or the other. It turns out that the interspecific transfer of mtDNA is asymetric: only wolves contained coyote genotypes, not the other way around. Since mtDNA is inherited only through the female line, and all hybrids are phenotypically indistinguishable from grey wolves, apparently only male wolves mate with female coyotes (if the opposite occurs, the hybrids are evidently unviable). Therefore, they can remain distinct species under the BSC - it is incomplete hybridization, especially since "wolf" appears dominant.So, we have narrowed it down a bit, anyway:1. Canis familiaris is an invalid taxon. Dogs are wolves.2. Canis rufus is a chimera, but breeds true. It is a valid taxon.3. Canis latrans remains a valid taxon, because F1 hybridization is asymetric. No pure coyote-wolf hybrid is viable to the point that a separate, distinct population of hybrids can exist (no wolf mtDNA shows up in coyotes).4. Canis lupus RULES!!!So Mr. Jack: How would you reconcile the N.A. canids under your schema?

I agree completely, I'm not a fan of the concept of species, I think it is a historical relic which only serves to confuse things. Sadly it is also a rather fundamental concept in too many branches of biology for it to be so glibly discarded.I think that if we need a definition of species then it has to be one suitable to the task at hand, and we shouldn't be worried if we come up with radically differing answers to what is and isn't a particular species in one context and not in another when we are asking fundamentally different questions, it certainly isn't a sign for the creationists to start rubbing their hands with glee.

Since ultimately biological evolution is the change of allele frequency over time, speciation would represent the change in allele frequency of a population such that it no longer shares the same allelic composition or frequency distribution of other similar populations and ceases to exchange genetic information with other populations...as you go to genus family etc virtually no alleles are shared. Incipient speciation would be the point in the process by which portions of a once interbreeding population begin to have a distinguishable allelic frequency. Thus, one would define species by taking population genetic measures as opposed to looking at individuals and seeing if they can mate i.e. chihuahau and rotweiler..

Hi again Quetzal, esquire. If you wish to use BSC as a working model, and not a strict definition, I have no problem with that. Thank you for your enlightening description of how it can be practically applied. I would clasify all four as species.For those you diferentiate as species I accept all your reasons, and merely add further support based on morphological and behavioural distinctions.Canus familiarus is distinct in behaviour and physiology from a wolf. For example a typical dog's bite actually has less force behind it than a human bite (although those sharp teeth make it rather more dangerous), while a wolf bites several times as hard. There are the obvious visual differences in shape and form, and it is generally acknowledged that wolves, and wolf-dog hybrids, are difficult to train and dangerous to keep as pets - although this could also be said of certain fighting dog breeds.Although I believe a strong case can be made for subdividing canus familiarus at least into subspecies, if not into actual species - my, admittedly limited, knowledge of dog breeds would lead me to conclude that they hold too continous a graduation of forms for any clear dividing lines to be drawn. It should be noted that all recognised breeds 'breed true' in that offspring of two (say) border collies is always itself a recognisable border collie.I am puzzled by your comments on C. rufus, I had not come across 'breeding true' at any point in the definition of BSC. It is not based on reproductive isolation, not whether the offspring of a group retain the characteristics of that group?

Snaps for MrHambre, I am in complete agreement.

I have no problem with your definition of the terms species and speciation. Neither I nor Wounded King is asserting that species don't exist or any such nonsense. It's just a fact that our notion of a species is based on a general analysis of shared alleles within a population and in the context of other populations, not some magic line that separates two essentially distinct kinds. Similarly, speciation is a process that does eventually result in genetically independent populations, but the magic threshold where the old subpopulation becomes the new species is, let's face it, arbitrary.And this is exactly the kind of confusion we'd expect from Darwinian evolution, as Wounded King noted. According to the creation model there should be only minor variation within a species and an absolutely clear demarcation between species. That we have to debate so furiously over how to recognize a new species is a testament to the incredible diversity produced by the Darwinian mechanism, even within an ostensibly homogeneous category.{edited second paragraph for clarity}
------------------
En la tierra de ciegos, el tuerto es el Rey.[This message has been edited by MrHambre, 08-13-2003]

Hi Mr.H,
I think we are all on the same page roughly. I just noticed that the examples tended to focus on indivdual breeding scenarios like one dog breed with another. My point was that since speciation is largely a population genetic issue one will not get far looking at discrete events. Thus, even if L.africana can successfully produce offspring with E. maximus, they are still separate species as they have wildly divergent allelic frequencies, only one recorded hybrid has ever been produced, in the wild they do not overlap in range, and it is probable they would exclude one another from their range by competing for the same resources...so while the discrete event that an F1 has been produced does not invalidate their taxanomic separation...and they are different genera! not even as close as species

Hey Mr. J: I'm not sure I agree with this. Oh, I mean in general I agree, but the specifics appear to conflict with what I've read elsewhere. For instance, feral dog packs exhibit increasingly "wolf-like" behavior over time, including identical pack and dominance behavior, territorialism, aggression, nocturnal preference, and even hunting strategies. Before I get too carried away with my comparison, let me clarify that what I mean by "wolf-like": they act just like other wild, social canids. It would be just as accurate to describe them as "coyote-like". The primary difference being that feral dogs tend to attack the throat (like wolves) rather than generically (like coyotes) although this is apparently not diagnostic. But I digress (see, for example, Gipson PS, Green JS 1994, "Feral Dogs", from USDA Publication "Prevention and Control of Wildlife Damage - 1994", pg C77-C81)Do you have any references for your bite comments? It would appear to me that the larger breeds of dogs, at least, have proportionally stronger jaws and bite capability. It's certainly well documented that feral dogs have been known to bring down deer, goats, and sheep, which would tend to imply they have pretty good jaw pressure. Even here, tho', you can't make a definitive comparison because coyotes have also been documented bringing down large prey and they are a much smaller critter than either the feral dog mongrels or wolves.Now where you might conceivably make a case is that self-sustaining populations of feral dogs should be considered either a distinct species or an incipient species. BSC breaks down a bit in this case (remember what I said about canid taxonomy being a major headache?). Unfortunately, there doesn't appear to be much homogeneity among separated populations - except that after X generations most of the stable packs in the US that have been studied tend to most closely resemble mongrelized german shephards and are fairly homogenous. A significant exception is the so-called Carolina Dog, which was recently recognized as an official "Rare Breed" by the UKC. However, there's nearly as much diversity in wild populations as there is between domestic dog breeds. Consider the Galapagos feral dogs (who represent a major conservation problem on the islands), which bear resemblance to the borzoi and other breeds that were the founder populations. So classifying them as a species would be problematic under the BSC. In addition, there are no known self sustaining feral dog-wolf or -coyote hybrid populations (in spite of urban legend to the contrary). In those areas where feral dog packs have established themselves, they exercise a fairly effective competitive exclusion on both wolves and coyotes. The known dog-other canid hybrids are evidently first generation between wild canids and free-ranging (vice truly feral) dogs. (Note: I'm not convinced this has been compellingly demonstrated.)As to the hybrids being dangerous to train - well, yeah, that would follow. You're crossing a domesticated breed with a non-domesticated wild animal that hasn't had hundreds of generations of "pacifying" artificial selection. That's a volitile mixture, to say the least. Breeding true isn't necessarily the case. I'm certainly no expert on dog breeding, but it's my understanding that even in well-established lineages there are often sports and atavisms, and that breeders are constantly attempting to maintain the purity of bloodlines (artificial selection, again). Anecdotally, my darling Pumpkin, an allegedly pure-bred golden retriever who proudly holds AKC papers attesting to eight generations of pure, championship golden retriever blood, is probably the furthest thing you can imagine from looking like a golden retriever (which was why the breeder wanted to dump her or euthanize her before we convinced him to part with her). In fact, she looks exactly like an Irish setter, down to ears, coat color, skinny hips, etc. So yeah, she's still a "golden retriever", but she doesn't LOOK like a goldie, and if you didn't know her lineage, your first impression would be a setter. So at least in this case, your generic statement that a border collie is always a border collie, doesn't appear to be accurate. Maybe a breeder can tell us more about this, but it doesn't appear to lend weight to reclassifying domestic, artificially selected dogs into multiple species categories. It isn't, strictly. The way I used it however is a logical follow-on from "two organisms that produce fertile offspring". It takes into consideration the multi-generation effect of the reproductive isolation/speciation. A species needs to be persistent, in my book. rufus fits the bill. Contrast this with wolf-dog or wolf-coyote hybrids, even recognized subspecies like Canis lupus laycaon whose populations tend to homegenize over time into one or the other parent populations - wolves or coyotes. These hybrid populations would NOT be considered distinct species, because they aren't persistent: they don't breed true.Note also that, although the general concensus is that the red wolf is a chimera, this has been disputed by some scientists, who claim that rufus has a distinct lineage going back 500-700,000 years. Their explanation for the screwy genotype of modern red wolves is that it is "contamination" from other canids over the last few hundred years due to habitat degradation - fewer potential mates of your own species, you take what you can get. I'm not personally wedded to either idea, except that the chimera scenario appeals to my sense of the absurd.

Doing a google search I found a consistently quoted value of up to 1,500 psi for a wolf bite. One of the more readable sites is here. While dog bites seem to be quoted as 200 psi for a typical breed, with up to 450 psi for large dogs - reference. However my comment that dog bites are weaker than human bites is incorrect, this source gives the average human bite at 165 psi, with the highest ever recorded at an astonishing 975 psi.I think these figures clearly show a marked difference in dog and wolf bite capabilities. Yes, it seems I was over zealous in claiming all breeds bred true. This is certainly not the case. However, my cousins did breed working collies (they are welsh sheep farmers), and all their puppies were also border collies. Anecdotal evidence, and all that, I know.

As Mam and others have alluded to, the fact that a concrete definition of 'species' is hard, if not impossible, to construct is itself supportive, in a circumstantial way, of evolution in general.Asfor what it takes to get a new species, I think that has been addressed rather thoroughly.------------------
(2) "A second characteristic of the pseudo-scientist, which greatly strengthens his isolation, is a tendency toward paranoia," which manifests itself in several ways:
...(3) He believes himself unjustly persecuted and discriminated against...(4) He has strong compulsions to focus his attacks on the greatest scientists and the best-established theories. ..

Hi Mr. J: Thanks for the references. Interesting. It seems counterintuitive somehow, but that would certainly be an indicator that we're dealing with different species adapted to different diets and habitats. OTOH, we still come back to genetics, both the nearly identical genotypes and the tendancy for dog genomes to be swamped in hybrid populations. I think you'd have a stronger case if pushed to have dogs as a distinct species by focusing on the feral populations - the only true "wild dog" populations we've got to study.

I think it makes quite a lot of sense. Wolves are frikkin' dangerous creatures and none of the purposes we put dogs require as powerful bite as a wolf possesses. You can imagine breeding a less dangerous animal being quite high on the priority list of the early domesticators. How about looking at Dingos? They are believed to be descended from domesticated dogs, introduced some 15,000 years ago. Can't find much about them on the 'net though.

Yeah, that would be another interesting group to study. Should they be considered Canis lupus dingo, or granting dogs species status, should they be Canis familiaris dingo, or even a separate species of their own like Canis dingo. I've actually seen all three categorizations. From what I've read, they have the same kind of descent with modification from wolves as other dogs do. Of course the entire wolf to dog diversification is complicated by the fact that there appears to have been at least four separate domestication events. They are definitely self-sustaining wild populations, don't appear to be tremendously interested in hybridizing with other domestic dogs, etc.