Dave,
The retroviral infection would have occurred in only one member of the species, true? Then it would have spread through the population via reproduction. Today, do all members of the species carry it? If so, why? Wouldn't there be lineages that carried it and others that don't?
Awesome questions. I had been thinking about those as well. Disclaimer: The following is what I think is the answer, but I need to do more research.
Yes, the retroviral infection happens in a single member of the species. The infection occurs at a specific point in the creature's genome, on a specific chromosome. Recall that half of an offspring's gene's are contributed by the mother, and half by the father, so this infection will land at a particular spot in the offspring that will be either a male-contributed chromosome or a female-contributed chromosome. If the infection first occurred in a female, the infection would land in a spot in the offspring that is passed on only if the offspring is female. In other words, the infection would be passed through females only - a matriarchal succession of infection. (This is the part I need to double-check on. Perhaps those with more knowledge of genetics can help me here.)
As the pyramid of infection grows, soon all members of a society will have it, because it is a permanent infection that can never be "healed." However, a geographically or reproductively isolated group of the species would not have the infection.
Therefore, it is possible for two individuals to be related but not share the infection, BUT any two individuals that do share it are definitely related. We might consider it as an extremely specific disorder that runs only in your family. It would be possible for you to have a sibling that is spared from it, but if you meet some other random person that has it, they are definitely related to you.
It is possible that there will be members of a species that never receive the infection, but that would only be because of reproductive isolation, possibly because they are geographically separated from the infected group. And it is precisely this type of isolation that leads to evolutionary branches. So the two groups will eventually diverge. If they meet up again somehow before a divergence happens, then the infection will spread to the uninfected group.
Just to clarify, let's consider a scenario. Say we had a group of cats, all of whom share a base retroviral infection. We'll call that one Infection 1. One day, a hapless female gets her ovum infected with Infection 2. This ovum becomes a kitten who now carries both infections. If the kitten is a male, then Infection 2 is not passed on any further, because it is in a female-contributed location on the genome, which a male cat cannot contribute. But if the kitten is a female, then all of her offspring will have the infection. So the infection begins to spread.
However, part of this group lives in the Eastern US and half in the Wastern US, and no interbreeding happens because the kitties can't make it over the Rockies. Poor things just don't like the cold. In that case, Infection 2 would be limited to, say, the Western cats. So at this point in time, we have two groups: Infection 1 occurs in all members of the species, showing that they are all related. Infection 2 occurs in a sub-group of the species only, showing that this group is more closely related than the species in general.
If the groups remain separated, in time they will diverge and reproductive isolation will be strengthened so that even if they were to meet up, they would not interbreed. (A lion will not mate with a housecat.) The infections would, however, be a record of their varying degrees of relatedness.
To recap, the only thing I'm unsure about is whether a retroviral infection actually travels down a gender-specific line as I have guessed. But this is actually immaterial to the bulk of the discussion. For the retroviral infection to have significance in tagging related species, the only thing necessary is that it is passed on through reproduction.
SNG