Various ideas that may or may not become part of a story.
Fate of a given mutation: I have read that a random mutation that has neither positive nor negative effect on survival rate will either dwindle to zero or fixate on the whole population, but of course that takes many many generations – hundreds? thousands? If a mutation negatively impacts survival, it is much more likely to dwindle to zero, and if it positively impacts survival, it is more likely to fixate.
Geographical or other variations: Consider a mutation like sickle-cell anemia: it has a negative effect on survival, most of the time, but for the part of the population living in mosquito-infested swamp areas, those members of the population with sickle-cell anemia are more likely to survive an outbreak of malaria, so the impact is positive in those areas. Thus the mutation survives in a halfway-state, never able to fixate all of the population, but never completely going extinct.
Dominant/recessive: Recessive traits are those that don’t show up in an individual unless the gene carrying them is “doubled up” with another gene like it. If the gene is paired with a non-matching gene, the dominant one will manifest itself and the recessive one will merely by “carried” in the genome unobserved.
How a recessive trait may appear: Children inherit one half (randomly) from each parent and combine the two, so if one parent is a carrier and has one half of the gene, and the other parent is not, two of four children will have the recessive half, the other two will be normal non-carriers. If BOTH parents are carriers, then two of the four children will also be carriers, one will be normal, and one will have the double-gene necessary to actually show the trait. If a double-carrier mates with a non-carrier, all four children will be single-carriers. If a double-carrier mates with a single-carrier, two children will be single-carriers and two will be double-carriers.
Delayed effects: The idea that a mutation may occur in a single member without being really noticed by the original mutant/carrier or her peers. The mutation spreads through the population by random chance over many generations. For example, a mutant might have latent psychic abilities, or latent immunity to a future disease or environmental effect. Here is an article about tetra-chromatic vision, or the ability to distringuish colors that appear identical to most humans, which is an interesting example of a possible mutation that might go undetected for some time. This is also a great idea for a story because the mutation might grow to a large fraction of the population before it is detected, and this leads to weird effects on people as they start to realize which families have the mutation and which do not. Chances of finding the original mutant individual would be slim after hundreds of generations, so the cause-and-effect is a bit more mysterious.
Sex-linked traits: Sex-linked traits are traits in either the X or Y chromosome, which have radically different outcomes if the individual is male (XY) or female (XX). If a trait is linked to the Y chromosome, then only men have it and it can only be passed to sons, not daughters. If a trait is linked to the X chromosome, then either men or women may carry it, and only women will be eligible to have a double-dose of it. Since men only have one X chromosome, it doesn’t matter if the trait is dominant or recessive, there is only one so it will always be expressed.
Hemophilia example: For example, hemophilia is a genetic disease that is sex-linked. Females may be carriers but do not normally suffer from the disease, since they are more likely to have one normal gene and one hemophilia gene. If a woman is a carrier, her children are equally likely to be one of these four: a carrier female, a normal female, a normal male, and a hemophiliac male. If the hemophiliac male has children, they are equally likely to be the same four types. If a female carrier and a hemophiliac male should happen to mate, their children would be of three types: hemophiliac male (50%), carrier female (25%) and hemophiliac female (25%). If a female has both genes for the trait, she will have the disease, and depending on the different variations of the disease may have abnormally heavy or prolonged menstrual periods leading to anemia/iron deficiency.
Reproduction by proxy: The idea that a particular member of a family may not breed, but will help the breeding members of the family to survive better. Observed on talkorigins.org: In some cases, organisms will completely forgo reproducing and only help their relatives reproduce. Ants, and other eusocial insects, have sterile castes that only serve the queen and assist her reproductive efforts. The sterile workers are reproducing by proxy. Therefore it doesn’t matter what genetic makeup the non-breeding members have, but if the breeding members are more likely to produce “better” workers, and the workers are more effective at helping the whole family survive, then there is an increased chance that genetic line will become dominant.
Combining an advantage with a disadvantage: Not a science idea, but a story idea. For better balance it might be interesting to combine an advantage with a disadvantage. For example, mutants that get a great new beneficial trait might also get a disadvantage along with it, which either makes their genetic line less likely to survive, or forces them to make an alliance with the non-mutant population that is symbiotic in nature.
Formation of a species: Unfortunately Darwin’s original work on evolution, On the Origin of Species, actually is kind of vague on how a species is really formed. =) It was more focused on what causes variation within a species, but when it comes time to form a new one, the details are hazy. In order to form a new species, the members of the new bloodline have to become different enough such that they could not interbreed with the old bloodlines even if they tried. Mostly this is due to the population being split up geographically for a long period of time, and even if they are brought back together later, they are too different to interbreed. However, populations can become “isolated” from each other by their behavior, instead of geography, such as when some insects switch to a different host plant, so that only certain members of the population are available for breeding, others are not.
As observed in nature, it appears that new species are only formed when there is some kind of isolation. This suggests a gradual change over hundreds or thousands of generations, where one generation and the next are completely compatible, but an individual brought forward from the past would probably not be compatible and would be unable to breed with the later generations.
Anyway, it makes sense that at least two mutations have to occur for a new species to form. 1. The members of the new species have to be able to breed with one another or they will not survive, and 2. they have to be different enough from the previous generations to be called a new species. If one individual mutant is different enough to be incompatible with her peers, she is essentially a lone member of a new species which will quickly become extinct. Therefore, there must be one mutation that creates a new variation (B) within the orignial species (A), which is still compatible reproductively with the A group that doesn’t have the trait. Then (probably much later) a second change (C) will occur and the members of group C are no longer compatible with group A, but are still compatible with group B.
Since mutants are extremely rare, there may be no more individuals of group A left at all – that is, change B may have already fixated on the population (or at least the isolated part of the population, which takes much less time). Change C is likely to come along much later when there are no A individuals left surviving. But, what if the second change occurs while A and B are still living side-by-side? The new family line C could only produce offspring with B, and mating with group A would produce no offspring (or sterile offspring, such as a mule) Therefore there may be some interesting story lines there for explaining why the mutants in family C cannot reliably reproduce and how they deal with this. Depending on how widespread the B’s are among the A population, this might make it less likely for the C family line to survive. Also, having a sterile offspring is probably worse for the survival of the bloodline than having no conception from the mating, because the sterile individual consumes family resources and doesn’t contribute to the propagation of the bloodline. (Interestingly, this article says that male mules are very sexually active, even though they are sterile, so are usually gelded to make them more manageable).
That’s all for now…
I think that most mutations radical enough to make the bearer no longer interfertile with its parent species also make it just plain dead, but a change in the number of chromosomes is an obvious counterexample (IIRC, humans and chimpanzees, despite being quite closely related, have differing chromosome counts).
I expect that’s probably another way to get the isolation necessary for speciation; having different numbers of chromosomes prevents interbreeding. And it must be fairly easy (for some definition of the word) for chromosome number changes to match up, or the chimp-human difference couldn’t reasonably have happened.
Again IIRC, the number of generations required for a gene to be weeded out of a population, even if it’s bad (lethal recessive) goes as the square of the population size; I expect a similar scaling law applies for a good gene spreading throughout the entire population, which is why you need isolation of a small group for speciation.
My understanding is that very few single mutations are fundamental enough to make a new species; even after the isolating event (if there is a single event), the two groups will be similar enough that a biologist from the pre-DNA-sequencing age would consider them the same species until different selection pressures have acted on them for evolutionarily significant times, by which time they will be separated by hundreds or thousands of mutations.
What you call “reproduction by proxy” is, I think, called “kin selection” in Biologese.
As you know, Bob, one of the best ways for the well-read layman to learn about evolutionary biology is to read the works of the late Stephen J Gould.
The Mendelian dominant/recessive scheme everyone learns about in school is the simplest possibility; there are all sorts of exciting things with partial or mixed dominance (eg, blood types), multiple genes affecting one trait, etc. (And, really, almost all genes affect multiple phenotypic traits to some degree, since down at that level it’s all proteins.) For an example of how complicated this can get, look up the genetics of dog coat coloration.
Oh, and don’t forget that in some sense, a lot of genes only determine potential; the environment that the organism develops in determines which potentials are fulfilled, and to what extent…