There is a third point to consider, which is the rate of dispersion of a recessive trait into a large general population. Rates of dispersion have been worked out for a number of genetic traits, mostly by physical anthropologists attempting to work out the ancestry of human species. They are also used by geneticists working out the history of genetic disease in both human and domesticated species. The problem is the only way for the hypothesis that cystinuria is widespread in ferrets, only masked by diet, to be true is for the disease to have become entrenched in ferrets between 50 and 150 years ago; that is assuming all the ferrets remained breeders. The reason is because the trait is recessive, so there has to be a lot of it in the population for the disease to show up. Imagine a gumball machine holding a thousand pink gumballs and two blue ones. If the gumballs are well mixed, what are the chances the two blue gumballs will come out of the lot at the same time? For cystinuria to be widespread, a lot of carriers have to be breeding with a lot of carriers, and they have to have been doing it for quite a long time. That means you will see a much higher rate of disease than we currently see. I could write for hours and falsify this hypothesis even more, but I only need a single falsification, and I have given at least two. If the disease were widespread, then it would be showing up in roughly the same rates as dogs. It is only showing up in a miniscule, statistically insignificant population, and that is not what you see in a recessive disease that is widespread within the population. Everything we see suggests that in ferrets cystinuria is a recent development in a small section of the population. Here's the thing about recessive traits, either autosomal or X-chromosome linked. Because the expression of those traits is mathematically predictable, you can use math equations to extrapolate the minimum and maximum numbers of carriers in a given population, based on expressed characteristics. For example, because we know albinism is an autosomal recessive trait, we could expect that for every 100 albinos, there may be as many as 200 hidden carriers of the albino gene, simply based on Mendel's patterns of inheritance. This is because < will not carry the gene, will carry one gene, and < will be carry two and be an albino. The real numbers are not as simple to deduce because of a number of factors, but even if only one ferret out of a hundred with the trait showed symptoms of cystinuria, if it was widespread in the population, there would be more sick ferrets than we are seeing. Let me put it another way. This entire argument rests on a single question: can ferrets that have undiagnosed cystinuria consume a 30-35% protein diet without forming stones? This question HAS to be true if cystinuria were widespread and never seen. But we know that the precipitation cascade in animals with cystinuria forms stones when the conditions are right, regardless of the protein levels of the food. If the disease were widespread, SOME ferrets would get dehydrated and form stones, SOME would have acidic urine and make them, SOME would eat too much protein and make them, SOME would eat too much salt and make them, and SOME would just make them for no real reason. There might be dozens of other factors that would result in the precipitation of cystine. These are real, current factors, not hypothetical. So, where are the sick ferrets? You can't have it both ways. If cystinuria was widespread, then more ferrets would show the disease. If it is not widespread, well, that explains why more ferrets don't show the disease. Q: "But I did want to point out what seems like a hole in your logic -- if ferrets are indeed domesticated, then the selective pressures are those of the conditions humans impose, and not those acting on wild weasels. You did say 'cause stones in a normal, average ferret', and I am talking about selective pressure that would allow toleration of non-meat feed." I am so glad you picked up on this, because it is a great question and incisive point, one greatly discussed in zooarchaeology. The answer is, ALL domesticated animals, even those under human selection, are still under natural selection. Humans select for color, nature selects for resistance to disease. You cannot remove an animal from natural selection, not even humans. We are in a closed system and natural selection is one of the rules under which the system operates. Aside from that, can you breed animals where the breeding process results in changes to basic, "core" physiology? [Posted in FML issue 4789]