transgenerational epigenetic inheritance
Could what your grandmother ate or endured leave a mark in your genes — not in the DNA letters, but in the chemical tags on top of them — that you then carry and pass on? This is one of the most exciting and most over-claimed ideas in modern biology. Transgenerational epigenetic inheritance is the proposal that epigenetic marks acquired in one generation are passed through the germ line to descendants who never experienced the original trigger.
To take the claim seriously you have to be strict about what 'transgenerational' means, because the body has a powerful eraser that stands in the way. Twice in the mammalian life cycle — once in the germ cells that will become eggs and sperm, and again just after fertilization — most epigenetic marks, including DNA methylation, are stripped off the genome and reset, a deliberate wipe that clears each generation's slate. Apparent inheritance can also be a trick of exposure: if a pregnant mother is exposed to something, her fetus is exposed too, and so are that fetus's own germ cells, which become her grandchildren — so an effect seen in children and grandchildren can be direct exposure, not true inheritance. Genuine transgenerational inheritance requires showing an effect in a generation that was never exposed even as a germ cell (typically the great-grandchildren on the female side), passed through a mark that survived both rounds of reprogramming.
Where does the honest line fall? In plants and in simple animals like the worm C. elegans, real transgenerational epigenetic inheritance is well documented — those organisms reprogram far less aggressively. In mammals, including humans, a few intriguing cases exist but rigorous, exposure-free, multi-generation proof is rare, and most headlines about 'inheriting trauma' or 'ancestral diet' run far ahead of the evidence. The safe summary: epigenetic inheritance across generations is real in some organisms, plausible but largely unproven in mammals, and routinely overstated in popular media. Treat dramatic human claims with caution, not because the idea is impossible, but because the bar of proof is high and seldom cleared.
In the worm C. elegans, a silencing signal triggered by a virus or by experimental RNA can be inherited for many generations without the DNA changing — a clean case of transgenerational epigenetic inheritance, in an animal that reprograms its genome far less thoroughly than mammals do.
Real in worms and plants; plausible but largely unproven in mammals — and routinely overstated in headlines.
The single biggest source of error is mistaking direct exposure for inheritance: a pregnant mother's exposure also hits the fetus and the fetus's germ cells. True transgenerational inheritance must show up in a generation never exposed even as a germ cell — a bar mammalian studies rarely meet.