In the winter of 1944, a German blockade cut off food to the cities of the western Netherlands. For roughly six months, until liberation in May 1945, millions of people in Amsterdam, Rotterdam and The Hague went hungry. Thousands starved. Women still conceived and carried children through it.
Those children became one of the most studied groups in medicine. Researchers in Amsterdam and Leiden have followed them for decades, and what they found is unsettling. People conceived during the famine had double the rate of coronary heart disease later in life. They were more likely to develop schizophrenia and depression, and they responded more strongly to stress. People exposed at any point in the womb had more type 2 diabetes.1 A few months of hunger before birth was still visible in their bodies sixty years later.
How does a body remember something that happened before it was born?
A mark on the genome
Part of the answer may be epigenetics. Your DNA sequence is the same in nearly every cell, but cells mark it with chemical tags that control which genes are switched on. The best studied of these tags is DNA methylation, a small chemical group attached to the DNA itself. Methylation doesn't change the sequence, but it changes how the sequence is read.
In 2008, Bas Heijmans and colleagues compared people who had been exposed to the famine in the womb with their own brothers and sisters who had not. Six decades later, the exposed siblings had less methylation at IGF2, an imprinted gene involved in growth. The difference showed up only in people exposed around the time of conception.2 It was the first direct evidence in humans that conditions in early life can leave epigenetic changes that last a lifetime. A later study found methylation differences at genes tied to growth and metabolism as well.3
From there it is a short and tempting leap. If famine can mark a person's genome, could those marks travel through sperm and eggs to their children, and their grandchildren? This idea is called transgenerational epigenetic inheritance. It would mean that what happens to you could shape the biology of descendants you never meet.
Three generations in one body
The Dutch researchers did look at the grandchildren. In 2013 they surveyed 360 adult children of people born around the time of the famine. The children of men who had been exposed in the womb were heavier, by about 1.6 points of body mass index on average. The children of exposed women showed no such difference, and neither group differed in adult health.4
It is an intriguing result, but it can't settle the question, and the reason is one of the strangest facts in biology. When a pregnant woman goes hungry, three generations are exposed at once: the woman herself, the fetus she is carrying, and the germ cells already forming inside that fetus, which will one day become her grandchildren. A grandchild of a famine-exposed woman was, in a real sense, physically present during the famine.
That is why researchers draw a line between intergenerational effects, which come from direct exposure, and truly transgenerational ones. To show that an epigenetic mark was inherited rather than caused directly, you need a generation that was never exposed at all. After exposure in pregnancy, that means the great-grandchildren.5 Nobody has that data for the Hunger Winter.
The great erasure
There is also a biological barrier. When the body makes the cells that will become sperm and eggs, it strips most of the methylation off their DNA and starts again. In mice, this reset erases most marks, including the parental imprints, with only rare regions escaping.6 Human germ cells go through a similar reset.7 In 2024, Mitinori Saitou's group in Kyoto managed to push human germ cell precursors grown from stem cells through this reset in the lab, and showed that the enzyme TET1 is needed for it to finish.8 That gives researchers a way to test directly which marks, if any, survive.
For an acquired mark to reach a grandchild, it would have to slip through this erasure, and then through a second wave of reprogramming in the early embryo. It is not impossible. But it is a narrow gate.
What the mice say
The strongest evidence comes from animals, where scientists can control the exposure and count the generations. In a 2014 study, Brian Dias and Kerry Ressler trained male mice to fear the smell of acetophenone, which smells a little like cherries. Their offspring, and their offspring's offspring, were more sensitive to that particular smell, even though they had never encountered it. The fathers' sperm carried less methylation at the gene for the receptor that detects it, and the effect held up through IVF and when pups were raised by unrelated mothers.9 Other work has pointed to small RNAs carried in sperm. Sperm RNA fragments from male mice on a high-fat diet were enough to cause metabolic problems in offspring.10
These results are striking, and they remain debated. Most stop at the second generation, which after paternal exposure is still the first one never directly exposed, and many have been hard to extend further.511
The children of survivors
The human evidence beyond famine is thinner. In 2016, Rachel Yehuda and colleagues measured methylation at FKBP5, a gene involved in the stress response, in 32 Holocaust survivors and 22 of their adult children, along with small control groups. Both survivors and their children showed differences at the same site, but in opposite directions: higher in the parents, lower in the children.12
It is a small study, and it can't separate biology from upbringing. Children of trauma survivors grow up with their parents' trauma. They share homes, stories, diets and stress. Any of those could shape a child's epigenome directly, with nothing passing through the germline at all.11
Where that leaves us
The reviews written by people who know this field best all land in the same place. Environmental exposures clearly leave epigenetic marks within a lifetime. Effects on the next generation are real, at least in the sense of direct exposure in the womb. But true transgenerational inheritance of acquired marks in humans has not been demonstrated, and in mammals generally the evidence remains inconclusive.51113
I don't find that disappointing. The children of the Hunger Winter already tell us something important: a mother's circumstances in the first weeks of pregnancy can shape her child's health for life. That is reason enough to care about the nutrition and wellbeing of people who are pregnant, or might become pregnant, wherever they live. Whether the marks travel one generation further is a fascinating scientific question. It doesn't change what we should do about the first.