On another forum there is a multi page train wreck being driven by a young earth creationist and an evolution denier, who are both steaming ahead into oblivion paying no regard to any arguement put to the two of them. Aside from facepalming, I am not part of the discussion.
However, something did occur to me. Someone tried the standard “we can breed dogs to have whatever trait we want in just a few generations” arguement to try to convince the evolution denier that genetic change over generations was possible… he failed.
I want to take that a step further (only as a mental exercise!)
If I want a dog “breed” that is the size of a Great Dane but has short legs, I need look no further than cross breeding a Dane with, say a Basset Hound and selecting puppies with short legs to breed the next generation. Apparently dalmatians took around 7 generations.
If I want an animal that is the size of a Great Dane etc but is unable to reproduce with a dog (ie, speciation at it’s most basic definition) how many generations might that take?
It happens in one generation all the time. Particularly in plants, it sometimes happens that diploid eggs and sperm, with a double set of chromosomes, are produced instead of the usual haploid gametes. If these combine, it will produce a tetraploid individual with four sets of chromosomes that will be unable to breed with the parental form, but can self-pollinate or breed with other tetraploids. Bingo, a new species by the biological species concept.
My guess is, more time than you’ve got, if your only tool is selection of body type. Wolves, dingos, coyotes, and jackals can all reproduce with dogs, and they are thousands of generations separated. They are all in the same genus (Canis), and many animals which are close enough genetically to be put in the same genus can reproduce successfully – often they are separated by ecological niche, mating behavior, or geography rather than reproductive compatibility. Canis members cannot interbreed with other canid genera like foxes, Japanese Raccoon Dogs, African Bush Dogs.
You can easily create a dog which is unable to reproduce, period. Those already exist, breeds so physically deformed they cannot mate, or cannot give birth, without AI and c-sections. English Bulldogs and Pugs are two, there are others.
In animals, similar genetic anomalies can produce new species in a single generation as well. Certain species of whiptail lizards are triploid, with three sets of chromosomes. It is thought that they originate through hybridization between other species that result in an offspring with two sets of chromosomes from one species and one from another. These lizards are unable to reproduce either with the parental species or even with each other due to problems in meiosis produced by the odd number of chromosomes. However, they do reproduce parthenogenetically; these species are entirely female.
Scientists have produced a new species of whiptail lizardby crossing a triploid all female species with a diploid species and obtained a hybrid with four sets of chromosomes. These go on to reproduce parthenogentically and constitute a new artificially produced species.
In theory, with modern genetic techniques, you could engineer two forms unable to interbreed in a single generation. Take two different genes, and engineer forms that are lethal in combination but have no effect individually. Fix one form in one genetic line, and the other in the other one. If the two lines hybridize, the two forms in combination will be lethal and all offspring will die. The two forms with then be intersterile.
Note that the Biological Species Concept does not require that two species not produce hybrids, or that the hybrids are infertile. As noted by Ulfreida, many species of Canis are interfertile. The BSC only requires that hybrids are not produced regularly in nature. Frequently the mechanisms keeping two species distinct are not genetic, but behavioral or ecological.
Since dogs are produced by artificial selection, the BSC does not apply to them. However, different breeds today are physically incapable of interbreeding with one another (at least without help from humans). The only reason they are considered the same species is that there is a continuous line of mutts between all the breeds. If you eliminated all other breeds, Great Danes and Chihuahuas would represent different species. (Although I’m sure someone will come up with a story about how their Pekinese managed to impregnate a St. Bernard.)
I don’t think that’s the conclusion to be reached. Given the existing range of genetic variation within dogs, I think that selective breeding could produce a breed of dog unable to breed with any other breed of dog within relatively few generations,even without generating new variants of genes. The thing is, that has never been the objective of a selective breeding program, for obvious reasons.
Colibri brings up a good point in Chihuahuas.
Dna proves they descend from the same line of wolves as other dogs.
They certainly wouldn’t interbreed with a wolf naturally and the only thing that’s kept them linked is that can interbreed with other descendants like fennec Fox’s. There’s no argument that fennec foxes and wolves are different species.
We wouldn’t have coywolves if not for domestic intermediaries.
This is why persistent evolutionary change typically takes 1 million years. You have this interspecies lattice.
According to this we are looking at as little as 14 generations for at least some reproductive barriers and around 100 to genetic incompatibility if we are doing the selecting. Without some anomaly.
There are a lot of examples of human induced evolution in animals already, that werent on purpose.
Poison immune mice, tuskless elephants, etc.
Here’s a wild species developed in two generations
Considered so because they only mate with their kind .
I read an article about selective breeding that said agricultural animals are about 20-30 standard deviations better at certain tasks (like chicken weight) than they were in the 1950s. So I’m guessing a generation is about 1 year, so that is about 60 generations of selective breeding at work.
Anyway, I wonder if by using that you can get a rough idea of how long it takes to change a species via selective breeding. Does that mean every 2 generations of selective breeding you bring a trait 1 standard deviation to the right? So that after 6 generations, a trait that was originally only shared by 0.3% of the animals is now shared by 50%? I’m not sure.
Most hybrids are either sterile, or are interfertile with both parental species. Therefore they either don’t have descendants, or else backcross with the parents and disappear as a distinct lineage. It’s unusual for hybrids to be so choosy as to only choose fellow hybrids as mates. (Obviously, their parents were not so choosy.) It will be interesting to see how long this lineage persists.
It is now possible to induce reciprocal or ‘balanced’ translocations of chromosomes at targeted sites with modern genetic techniques. So, you can cause a break in, say, chromosome 6 and another break in chromosome 8 in such a way that the fragments could recombine into a chromosome 6/8 and a chromosome 8/6. The laboratory process is not particularly efficient. Individuals with such translocations are often healthy themselves but relatively infertile because their offspring may have the wrong complement of chromosomes: ie good copes of chromosomes 6 and 8 from the mother but a copy of 6/8 and 8 from the father, resulting in not enough genes from 6 and too many copies of genes from 8. This results in the embryo being nonviable.
If you took male and female fertilized eggs, and induced the same balanced translocation in both of them, then as adults they would be interfertile with each other but much less so with other members of the population. To be stable, there would need to be balanced translocations in both copies of the targeted chromosomes in each embryo.
It depends on the trait, and on the intensity of selection. For weight in chickens, you are talking about a large number of genes with multiple effects, so it may take a while. For trait like coat color, you could fix a recessive allele by simply breeding from individuals that showed the trait (and thus only had the recessive allele) in one generation. For dominant traits it might take longer to eliminate recessive traits if you could not do detailed genetic analysis.
True, but as interspecies hybrids, they are going to be starting out with more genetic variation than if you had that amount of inbreeding within species.