Leaving off from where I continued earlier, it’s now time for this …
And now it’s time for this bullshit to be shredded:
Ahem, over 1½ million peer reviewed scientific papers, document in exquisite detail, the evidence for evolution. This includes direct experimental test and verification of evolutionary postulates, and replication in the laboratory of speciation events.
And it’s at this point, that your scientific ignorance is starkly on display. Because, wait for it, Looby Loo, scientists don’t postulate that a population of organisms “turns into something else”, and they certainly don’t postulate the sort of farcical clade jumping that is a feature of the more preposterous creationist assertions.
Strap yourself in, you’re in for another hard ride.
Quite simply, a species, from the standpoint of biology, is nothing more than a population of organisms that can produce multiple generations of offspring, when the members thereof pair off and mate with each other. That is IT. Of course, taxonomists have a habit of confusing the issue with their approach to the matter, but the complications of that are properly the subject of another, and lengthy, dissertation.
Now, if a population of such organisms splits for some reason, then there is no reason why those two new populations should follow identical genetic trajectories. The workings of meiosis alone will make those trajectories diverge, and once mutations are added to the equation, that divergence becomes even greater. Among the genes that can diverge, once gene flow between the new populations ceases, are the fertillin genes, that are responsible for maintaining egg and sperm compatibility. Those two populations will eventually reach the point where some individuals of population A will be unable to produce fertile offspring with individuals of population B, and vice versa. The proportion thereof in each population will increase over time, until eventually, both populations are completely reproductively incompatible, at which point, we have a fully-fledged speciation event.
The point here being, that no one regards those species as “a whole new kind of organism” in the naive sense you’re peddling above. Not least, because those two new species manifestly share a common ancestor, and the clade within which they reside is a subclade of the clade of their ancestors. Indeed, Linnaeus embedded both clade nesting of this sort, and the relatedness of organisms in the biosphere, in his taxonomic system, and did so fully 62 years before Darwin was born. He treated relatedness of organisms (on the basis of comparative anatomy, a discipline he did much to add rigour to) as a useful brute fact, and left it to Darwin to devise an explanation for that brute fact.
For example, let’s imagine a scenario in which all the world’s cat species become extinct. After this somewhat alarming event, some dog species start moving into the requisite niches, and over time, start acquiring anatomical features of the sort that were previously seen in the now extinct cats. The point here is that these new dogs have NOT “turned into cats”, they are still dogs, courtesy of their ancestry, but have merely acquired features that result in them resembling the cats of old. Properly, they are cat-like dogs. So kindly drop the fatuous clade jumping idea you picked up from creationist propaganda, because it’s bullshit.
Indeed, it’s time for this:
Why The Business Of Defining ‘Species’ As A Class Entity Is Interesting - Part One
First of all, “species” is a dynamic concept. A distinction needs to be drawn between the taxonomic concept of species, which is a snapshot sample of a given population of organisms at a given moment in the history of the population for the purposes of cataloguing, and the actual reproducing population thus sampled. Now it so happens that for the majority of species that are thus sampled for taxonomic purposes, gross changes are not going to occur in a few generations (though for some fecund species with rapid reproduction cycles, cumulative changes over time are more noticeable than in less fecund species with slow reproduction cycles). Consequently, the ‘snapshot’ that is taken is likely to remain a valid reference point for that species for a considerable period of time, contributing to the illusion of fixity. However, as Streelman et al in their paper on Cynotilapia afra, a Cichlid fish from Lake Malawi that has been extensively studied, have already pointed out, we’re about to see this situation change in a vertebrate eukaryote. For reference, the paper in question is this one:
Hybridisation and Contemporary Evolution in an introduced Cichlid Fish from Lake Malawi National Park by J. Todd Streelman, S.L. Gymrek, M.R. Kidd, C. Kidd, R.L. Robinson, E. Hert, A.J. Ambali and T.D. Kocher, Molecular Ecology, 13: 2471-2479 (21 April 2004)
Basically, Cynotilapia afra was originally endemic to one restricted location in Lake Malawi. It belongs to that group of Cichlid fishes known as the Mbuna or Rockfish, which all exhibit characteristic behaviour patterns both in the wild and in an aquarium. They remain close to the rock screes that are abundant in Lake Malawi, and base their territories upon those rock screes. However, these rock screes are divided into isolated formations separated by wide stretches of open water, with open stretches of sandy bottom devoid of cover in the shallower parts of the lake. Mbuna do not cross from one scree to another across open water for good reason - the open water stretches are patrolled by other, predatory Cichlids such as Nimbochromis venustus that would make a meal of any Mbuna that strayed too far from the rocks. Over time, the Mbuna have become increasingly behaviourally tied to the rocks, and rarely stray more than a few metres from their ‘home patch’. This results in the appearance of isolated populations of Mbuna that hardly ever interact with each other, and thus the scene is set for speciation via isolation.
Now, because these fishes are valuable to the aquarium trade, and because some of the fishes in question are in locations that are troublesome for collectors to visit for various reasons (in the past these have included civil wars and other instances of political strife) one enterprising collector decided to take a number of Cynotilapia afra in the 1960s and relocate them to a less troublesome part of the lake. Thus the scene was set for an interesting biological experiment.
The fishes that were relocated set about breeding in their new location, and spreading slowly across the stretches of rock scree available to them, which happened to encircle a small island in the lake. The spread was relatively slow, but once it became entrenched in the early 1980s, scientists noticed that widely separated populations around this island were diverging from each other. Thus, the fishes were sampled on a regular basis for DNA in order to track the changes. Now, Streelman et al cited in their paper that these fishes are diverging at such a rate that they could provide the first documented example in the near future of a speciation event with a genetic audit trail. Which is going to be interesting not only because it will provide many useful insights into speciation mechanisms at the genetic level and inform us much more thoroughly about what constitutes ‘specieshood’ from a genetic standpoint (though already, scientists consider that major histocompatibility complex genes, which are already implicated in ‘self’ versus ‘nonself’ identity from an immune standpoint, will be implicated here), but also because it will bring the taxonomic problem into the limelight. It will also have an impact on cladistics, because interfertility tests with the original population could yield some interesting results once speciation has been documented to take place.
I shall illustrate this as follows. Let our original population from which the stock was taken prior to relocation (and the source of the original taxonomic specimens used to describe the species) be population A. The transplanted fishes became population B, which then diverged into populations B1 and B2, with decreasing mating interaction between them. Now, let us jump to the future, and arrive at the point where the fishes in populations B1 and B2 fail to be interfertile, and thus constitute new species. Four possibilities are now open and available for experimental testing:
[1] Individuals from populations B1 and B2 continue to be interfertile with individuals from population A;
[2] Individuals from population B1 fail to be interfertile with individuals from population A, but individuals from population B2 continue to be interfertile with individuals from population A;
[3] Individuals from population B2 fail to be interfertile with individuals from population A, but individuals from population B1 continue to be interfertile with individuals from population A;
[4] Individuals from populations B1 and B2 fail to be interfertile with individuals from population A.
If experiment should establish option that [1] holds true, we’re in a bit of a pickle. Because the fishes from populations B1 and B2 are still interfertile with the fishes from population A, but not interfertile with each other. Here, we would have to treat them as being ‘ring species’, and devise suitable new taxa to accommodate this.
If [2] or [3] hold, we’re in less trouble. The population that continues to be interfertile with A can keep the name Cynotilapia afra and we can erect a new species identity for the other. However, it’s still a conundrum for the cladistic purist, because the original Cynotilapia afra population is still extant and reproducing happily. It hasn’t undergone the ‘pseudo-extinction’ of splitting into two new species that is the more usual case considered.
If [4] holds, then we will have to erect new species identities for both populations B1 and B2, and again, the cladistic purist will have a conundrum to address because the original population A will still be extant. Worse still, the original population A will still be extant, AND we will have TWO new species to fit into a cladogram.
All of which demonstrates that our concept of ‘species’ will be in serious need of refinement once the appropriate real world results are in.
This is going to be fun when it happens, and I can hardly wait. I probably won’t live long enough to see it happen, but some of the other posters here may well do. I hope you all remember what I’ve posted above when it does happen! 
Meanwhile, it’s also time to expand upon the above, as follows:
Why The Business Of Defining ‘Species’ As A Class Entity Is Interesting - Part Two
As I’ve mentioned above, in any population of organisms, variation will be disseminated from one generation to another even before we factor in the appearance of mutations. This variation will be disseminated with respect to every gene in the constituent genomes of the parent organisms - the offspring will inherit different combinations thereof even if they share the same parents, let alone the differences that will be present between two organisms that have different parents in the population.
Now, even if we assume relatively stable environmental conditions, with no large-scale changes occurring over time, that environment will have an effect upon the inheritance patterns of some of those genes. Many genes will be neutral, and will themselves not determine competence with respect to survival and reproductive fitness. Other genes will have such an effect, and will thus be subject to the appropriate selection pressures. Genes resulting in reduced competence will eventually disappear over numerous generations, as the individuals possessing them produce fewer and fewer offspring, and genes resulting in increased competence will eventually dominate in the population, slowly but surely raising the bar of competence as it were.
Now, let us perform a little experiment. Let us sample some individuals at a given generation, say generation N, and by some suitable means, preserve their genomes so that these genomes can be resurrected at a later date. This is in effect what Richard Lenski did in his 20-year experiment with E. coli - he created what was in effect a genomic “fossil record” of past reproduction that could be resurrected and replayed at will in the future, though he did this on multiple occasions in order to track the generational point at which certain mutations appeared. My purpose with this thought experiment is related, but different in a key manner as shall be seen.
Now, let the organisms continue reproducing, through generation N+1, N+2, N+3 … all the way to, say, N+100,000. I choose this figure as being one that is likely to exhibit the requisite variation, but depending upon the organism chosen, the phenomenon I am about to describe could occur sooner. Now, after this extended period of reproduction and dissemination of variation across generations, one can ask the question “can the individuals in generation N+100,000 reproduce with their distant ancestors?”
So, we perform the requisite trial.
Now, it may happen that when we resurrect generation N, and mate individuals containing generation N genomes with individuals containing generation N+100,000 genomes, that mating is possible, and viable offspring result. It is also possible that this trial mating may FAIL to produce viable offspring, indeed it is possible that the generation N+100,000 individuals will never choose to mate with the generation N individuals if given a free choice, because those generation N+100,000 individuals recognise the generation N individuals as being substantively different in some way. The moment such a back-cross fails to result in compatible mating, we have a speciation event. The individuals of generation N+100,000 now, in effect, constitute a different species from those of generation N, because they can no longer mate with those ancestral individuals. Thus, the sampling that is performed for taxonomic purposes is necessarily a historical artefact, because it is entirely possible that after 100,000 generations have elapsed, the descendants of generation N (i.e., in this instance, generation N+100,000) will be reproductively incompatible with their generation N ancestors, and will in effect constitute a different species. The taxonomic sampling performed at generation N will therefore be a historical record of what that species once was, but a new sampling will have to take place in order to represent the species as it is now.
Taxonomic holotype specimens are chosen as “benchmark” specimens representing a given species in order to provide a reference standard for precise scientific work, but they are chosen on the understanding that they represent a particular generational sampling and that at some distant point in the future, they will fail to represent the species when the reproductive incompatibility described above takes place. A species is properly defined by reference to the reproducing population, and sampled individuals for taxonomic purposes do not constitute the species itself - they merely constitute a snapshot of that species at a particular instant for comparison purposes. The species itself is the reproducing population, and because that population undergoes changes with each new generation, it is a dynamic entity.
Of course, if the above experiment is ever performed, and that experiment results in reproductive incompatibility between generation N and generation N+100,000, despite the fact that generation N+100,000 is being back-crossed with with its own distant ancestors, then this will not only make a mockery of the idea that a species is a static entity (a certain creationist comes to mind here with respect to assertions about species being static), but it will mean that we have to think long and hard about how ‘species’ is defined rigorously. Because, if the above experiment is performed, and yields reproductive incompatibility between ancestors and descendants after a large number of generations, then according to the biological species concept, the two populations will constitute different species. But, here we will have two populations, A and B, that are reproductively incompatible, and hence deserving of separate and unique species identifiers, but population A will be the ancestors of population B. If this doesn’t tell you that a species is a dynamic entity, then nothing will. But, let’s move on …
Now, if we partition our generation N population into two separate populations, and keep them separate, then there is NO reason whatsoever to suggest that the two populations will produce identical versions of generation N+1, N+2, N+3 etc., precisely because of the dissemination of variation across generations. It is FAR more likely that the two separated populations will diverge. The genomes extant in Population 1 in a given future generation will be different from those extant in Population 2 in the same future generation. Allow a sufficient number of generations to elapse, and again, a trial mating between individuals of Population 1 and Population 2 could fail. When this happens, the two populations have become reproductively incompatible, and we have a speciation event. That’s what speciation means - populations becoming reproductively incompatible with other populations of the same ancestry.
And, lo and behold, this has been directly observed in the laboratory with various model organisms. Theodosius Dobzhansky produced such a speciation event in 1971 with a laboratory population of the fruit fly Drosophila pseudoobscura. After just five years in the laboratory, his isolated population was no longer able to produce viable offspring with the wild flies. The requisite scientific paper describing this process is this one:
Experimentally Created Incipient Species of Drosophila by Theodosius Dobzhansky & Olga Pavlovsky, Nature 230, pp 289 - 292 (2nd April 1971)
Other papers describe other speciation events generated in the laboratory, namely:
Evidence for rapid speciation following a founder event in the laboratory by J.R. Weinberg V. R. Starczak and P. Jora, Evolution 46:1214-1220 (1992)
Founder-flush speciation in Drosophila pseudoobscura: a large scale experiment by A. Galiana, A. Moya and F. J. Alaya, Evolution 47: 432-444 (1993)
In other words, speciation is an established fact, has been documented as such in the peer reviewed scientific literature, and indeed, there are other papers in the literature describing observed instances of speciation in the wild. A good one is the frog Hyla versicolor. This arose from an ancestral population of the frog Hyla chrysocelis. How do we know this? Because analysis of the genomes of the two species demonstrates that all of the genes of Hyla versicolor arose from Hyla chrysocelis. What makes Hyla versicolor different is that it has inherited two complete copies of the Hyla chrysocelis genome. Instead of being diploid, it is tetraploid - it has its chromosomes not in pairs, but in sets of four. It has, in effect, two entire sets of Hyla chrysocelis chromosomes. But, as a result of this, it can no longer reproduce with other frogs sharing the same ancestors. It can no longer reproduce with the diploid Hyla chrysocelis because triploid offspring (which would result from a fusion of gametes from one diploid and one tetraploid individual) are not viable. Again, this is established scientific fact, and no amount of wishful thinking can alter this.
Looks like you have at least three decades of scientific study to catch up on.