Mind & Body · Tuesday, 8 September 2026

01 Briefing what happened

Puberty is a brake coming off, not an engine starting, and the brake gene only works if it came from your father

The hormone system that drives puberty is already working in a baby. It is then held shut for about nine years, and puberty is the day the hold gives way.

3 months per decade

how much earlier girls' breasts start developing, 1977 to 2013

a meta-analysis of 30 studies puts the fall at 0.24 years each decade [3]

20-30%

of early puberty that runs in families, traced to one broken brake gene

MKRN3 also explains about 9 to 10 percent of cases with no family history [1]

50-80%

of the difference in when puberty starts, set by inherited genes

the rest is body fat, nutrition, sleep and things nobody has pinned down [1]

88.9 to 1,144.7

recorded early puberty per 100,000 South Korean girls, across two decades

Denmark's rate tripled in girls and doubled in boys between 1998 and 2017 [5]

The lead story — what happened

Puberty is not switched on. The hormone system that runs it already worked in the womb and in the first months after birth, and childhood is the years it spends held shut. [2]
The signal is a rhythm, not a level. The hypothalamus, the control room at the base of the brain, releases gonadotropin-releasing hormone in pulses, and those pulses tell the pituitary gland to wake the ovaries or testes. [2][24]
One gene does much of the holding. MKRN3 is at its highest in early childhood and falls sharply just before puberty begins, and it works by silencing the genes that make the starting signals. [1]
Children who inherit a broken copy of that gene start puberty years early. It is the commonest single-gene cause, behind 20 to 30 percent of early puberty that runs in families and about 9 to 10 percent of one-off cases. [1]
The broken copy only counts if it came from the father. Both parents pass MKRN3 on, but the mother's copy is switched off in every cell, so a fault in hers changes nothing. [1]
Genes set most of the date. Between 50 and 80 percent of the difference in when puberty starts is inherited. [1][33]
Body fat is a permission slip, not an engine. Leptin, the hormone fat cells release in proportion to how much fat there is, has to be present for puberty to begin, and going hungry delays it. [1][49]
The average date has been moving. Across 30 studies, the age at which girls' breasts start developing fell about 0.24 years, roughly three months, every decade between 1977 and 2013. [3]
Diagnoses have risen faster than that. In Denmark the recorded rate of early puberty doubled in boys and tripled in girls between 1998 and 2017. In South Korea the girls' rate went from 88.9 to 1,144.7 per 100,000. [5]
The line has not moved. Doctors still investigate breast development before age 8 in girls and testicle growth before age 9 in boys, thresholds drawn from an older population. [3][1]
Nobody knows what lets go. The Harvard review that set out this picture says plainly that the mechanisms behind the restart are not completely known. [2]

Who is involved

MKRN3

a gene switched on only in the copy inherited from the father; it holds the puberty signal down through childhood [1]
KNDy neurons

a small cluster in the hypothalamus making kisspeptin, the strongest known push on the puberty signal, alongside one chemical that starts each pulse and one that stops it [1][4]
Leptin

the hormone fat cells release in proportion to body fat; it is the permission signal the brain waits for before starting [1][49]
Ana Paula Abreu and Ursula Kaiser

hormone researchers at Brigham and Women's Hospital and Harvard Medical School; their review set out the hold-and-release picture [2]
Anders Juul

a children's hormone specialist at the University of Copenhagen who has studied puberty for over 30 years; he argues male puberty is badly under-researched [33]

How it unfolded

In the womb the puberty hormone system is already active [2]
First months of life it runs briefly again, then goes quiet [1][2]
Ages 1 to 8 the brake gene MKRN3 sits at its highest [1]
1977 to 2013 the average age of breast development falls about three months a decade [3]
2013 broken MKRN3 found in families with early puberty, by reading whole exomes [1]
Just before puberty MKRN3 activity falls sharply and the pulses return [1]

Where this points

Watch whether the age thresholds get redrawn: 8 and 9 were set from an older population, and the rising diagnosis counts in Denmark and South Korea are measured against them. [3][5]

What is pushing on the whole day

The bar and the word are our reading of how hard each one is pushing today. The arrow is where it is heading. The evidence is in the stories below.

The childhood hold Easing↓

the brake gene's activity drops sharply just before puberty starts [1]

Body fat in children Building↑

girls with obesity had early puberty at 11.8 percent against 3.2 percent of girls overall in two Chinese cities [5]

Chemicals that act like hormones Building↑

a review of 87 studies links low-dose mixtures to early puberty, but 42 of them were animal or dish work [16]

The definition of too early High→

age 8 in girls and 9 in boys, unchanged while the population average slid [3]

The rest of the day

40 more stories on this beat.

Each with its own sources. None of these is a link to the story above.

02
Early puberty linked to psychotic-like experiences

Researchers followed 11,758 American children in the ABCD study, a project that scans and surveys the same young people for years, from about age 9.9 to 12.9. They separated two things usually mixed together. Timing is whether a child reaches puberty before or after their classmates. Tempo is how fast they move through it once it starts. Earlier timing in both boys and girls went with more psychotic-like experiences, such as hearing or seeing things others do not. Faster tempo in boys went with fewer. [6]

Why it matters — What the measurements track is timing relative to other children, not the calendar age. That matters because a falling average moves who counts as early.
03
Early first period, higher breast cancer odds

A meta-analysis pooled 74 case-control studies published between January 2014 and February 2025 on the age of a girl's first period and her later breast cancer risk. Girls whose periods started before 13 had odds about 15 percent higher than those who started later. Before 12, the odds were about 27 percent higher. Women who started at 13 or older had lower odds than average. The effect varied by region and was largest in West Asia. [7]

Why it matters — More years of oestrogen is the mechanism usually given for this link. It is a modest effect on one woman and a large one across a population whose average age of first period keeps falling.
04
Puberty's timing reshapes the growing brain

A second analysis of the ABCD study used repeat brain scans to separate what puberty does from what simply getting older does. Children whose puberty came earlier than their peers showed brain development running ahead of the normal curve for their age. Children who moved through puberty faster had thicker cortex, more surface area and larger deep-brain structures than slower peers. Chronological age on its own did not predict these patterns. [8]

Why it matters — The brain's schedule is tied to the hormones, not to the birthday. Anything that shifts the age of puberty shifts the brain's timetable with it.
05
The brain has five eras, and adolescence runs to 32

Scientists at Cambridge University mapped the wiring of nearly 4,000 brains, from under one year old to 90, and found four turning points, at roughly ages 9, 32, 66 and 83. The childhood phase runs from birth to about nine, which is when puberty typically begins. The next phase, which they call adolescence, lasts on average until the early thirties. Alexa Mousley, who led the work, said it describes the pattern of change, not that people in their late twenties behave like teenagers. [9]

Why it matters — The first turning point lands on the age puberty starts. It also stretches what adolescence means far past the age most legal systems use for it.
06
Prefrontal chemistry rebalances through the teens

Using a very strong 7-tesla scanner, researchers tracked 153 people aged 10 to 32 across up to three visits, 272 in total. They measured glutamate, the brain's main go signal, and GABA, its main stop signal, in the patch of cortex behind the forehead used for holding things in mind. Early in adolescence the balance leaned towards go, and it shifted towards stop into adulthood. Iron in the deep brain, a stand-in for dopamine, tracked the change. [10]

Why it matters — It puts a chemical measurement under the familiar claim that teenage self-control is still being built. The years it covers are the years the hormones arrive.
07
Monkey study: wiring predicts, shrinking does not

Researchers tracked monkeys through adolescence, recording behaviour, the firing of individual neurons and repeated anatomical scans in the same animals. Working memory improved, and the trajectory of prefrontal activity tracked those gains closely. The improvements were predicted by the maturing of long white-matter tracts joining the frontal lobe to the rest of the brain. They were not predicted by the falling brain volume and thinning cortex that carry the story in humans. [11]

Why it matters — It is a direct challenge to the usual explanation of teenage brain development. If the connections matter more than the trimming, human studies built on brain volume may be watching the wrong part.
08
Two meta-analyses: heavier boys start earlier

One review pooled six high-quality studies covering 64,485 boys. Obesity raised the odds of early testicle growth by about 27 percent and of early pubic hair by 37 percent; being overweight raised the first by about 20 percent. A second review pooled 12 forward-looking cohorts up to July 2025 and found risk ratios of 1.38 for overweight boys and 1.43 for obese boys. The link had been treated as clear in girls and unsettled in boys. [12][13]

Why it matters — It closes the gap that let the obesity-and-puberty link be filed as a girls-only finding. Both reviews note the relationship may not hold at the very highest weights.
09
Early puberty counted in two Chinese cities

A forward-looking cohort followed 8,025 schoolchildren, 4,341 boys and 3,684 girls, in two Chinese cities. Adjusted incidence of precocious puberty was 1.7 percent overall, 2.9 percent in girls and 1.1 percent in boys. Early puberty ran at 3.9 percent. The split by weight was stark: 6.9 percent of girls with overweight and 11.8 percent of girls with obesity had early puberty, against 3.2 percent of girls overall. Urban boys were far ahead of suburban boys, 5.7 percent against 0.7. [5]

Why it matters — It is a population study rather than a count of clinic referrals. The urban-suburban gap inside one country is as wide as the gap between weight groups, and the paper cannot say why.
10
Leptin and insulin, the link to body fat

A review gathers the mechanisms connecting childhood obesity to earlier puberty in both sexes. The strongest candidate is leptin, the hormone released by fat cells in proportion to how much fat there is, acting on the kisspeptin neurons that push the puberty signal. Insulin, usually raised in obesity, adds to it. The gut's bacteria appear in the chain too, because they shape insulin resistance, which is itself linked to early puberty in girls. [14]

Why it matters — It explains how a change in what children eat reaches a switch inside the brain. It is also the reasoning behind the metformin trial further down this list.
11
Gut bacteria and the timing of puberty

A review gathered human and animal evidence that gut bacteria help set when puberty starts. Children with early puberty repeatedly show fewer bacteria that make short-chain fatty acids, such as Bacteroides, and more Alistipes, Ruminococcus and Lachnoclostridium. One proposed route is an enzyme made by gut bacteria that undoes the liver's packaging of oestrogen, so it gets absorbed again instead of leaving the body. In animals, disturbing the gut sped puberty up and treating it delayed the hormone axis. [15]

Why it matters — It puts a second brake somewhere nobody was looking: the gut rather than the brain. The step is removal being blocked, not production rising, which is the lead story's shape one layer further out.
12
Chemical mixtures and the gut-brain route

A systematic review pooled 87 studies, 45 in people, 32 in animals and 10 in dishes, on low-dose mixtures of hormone-mimicking chemicals and early puberty. Exposure around birth was consistently followed by a less varied gut community: Lactobacillus down 40 percent, Bacteroides up by half, and butyrate production halved. The authors argue that direct hormone disruption does not explain the effect on its own, and put the gut-brain link forward as the missing step. [16]

Why it matters — Nearly half the evidence is from animals or cell cultures, which the authors say themselves. It is a hypothesis with a mechanism attached, not a settled human finding.
13
Diet changed the brake gene's regulators

Rats fed high-fat or high-carbohydrate food early in life reached puberty sooner, shown by earlier vaginal opening, higher gonadotropins and oestradiol, and more mature ovaries. The researchers then looked at microRNAs in the hypothalamus, short molecules that turn other genes down, and found six had changed. Their targets included Mkrn3, the brake gene itself, and Kiss1, one of the push genes, moving in the opposite direction to the microRNAs. [17]

Why it matters — It is the clearest picture so far of how food reaches a switch in the brain. It also lands on exactly the gene the lead story is about.
14
Fat rats, early puberty, and a new peptide

Male rats raised from birth on food that was 45 percent fat were heavier and reached puberty earlier than rats on normal food. Their hypothalamus showed more activity in the genes for the starting hormone and for kisspeptin, and their blood carried more luteinising hormone and testosterone. A little-studied protein called phoenixin was raised in the hypothalamus, pituitary, testes and fat tissue, and stayed raised after adjusting for body size. [18]

Why it matters — It offers a candidate messenger between fat tissue and the brain in males, where the obesity link has been argued over. It is a rat study, and the protein has not been tested this way in people.
15
Heavy metals act on hormones too

A review sets out the evidence that cadmium, lead, mercury, arsenic and nickel interfere with the body's hormone signals. They come mostly from human activity and stay in soil, air and water. Long exposure, including at low doses, has been linked to diabetes, obesity, autoimmune thyroid disease, thyroid cancer, disorders of puberty and infertility, through oxidative stress, inflammation and interference with hormone receptors. How much damage they do depends on the dose, the duration and the chemical form. [19]

Why it matters — Metals are usually filed under poisoning rather than hormones. Adding them widens the list of exposures that could be moving the age of puberty.
16
Hormone-mimics and male fertility

A review of endocrine-disrupting chemicals and male fertility gathers the evidence on heavy metals, phthalates, pesticides and bisphenol A. They act by several routes at once: oxidative stress, cell death, interference with oestrogen and androgen signalling, disruption of the brain-to-testis hormone chain, and inherited changes in how genes are read. Markers of exposure in men track with lower sperm concentration, poorer movement and shape, and damaged sperm DNA. The authors argue regulation has not kept pace. [20]

Why it matters — It names the distance between what laboratory evidence shows and what any regulator currently requires. The same chemicals turn up in the early-puberty evidence elsewhere in this list.
17
Male breast growth and chemical exposure

Gynecomastia, breast tissue growing in men and boys, is the commonest non-cancerous breast condition in males, driven by an imbalance between oestrogen and androgen. A systematic review sets out which chemicals are implicated: bisphenols, phthalates and polycyclic aromatic hydrocarbons, which can mimic oestrogen, block androgens, or interfere with how hormones are broken down. Some evidence suggests the tissue also reacts with long-running, non-infectious inflammation, which may change how sensitive it is. [21]

Why it matters — It is a visible, countable effect of the same chemicals blamed for shifting the age of puberty. Boys going through puberty are one of the groups where it appears most.
18
Why children are the wrong body for these chemicals

Two reviews set out why hormone-mimicking chemicals hit children hardest. Their systems for breaking down toxins are unfinished, the barrier protecting the brain is still forming, and organs are still specialising, so exposure during those windows can leave marks that last. The second review follows the same chemicals across the placenta before birth. Reported links include delayed brain development, reproductive changes, obesity, type 2 diabetes and thyroid problems. [22][23]

Why it matters — It explains why the same air, water and food do different things at different ages. The window that matters most closes before a child is old enough to be asked anything.
19
Two percent of teenagers are simply late

Delayed puberty means the process starting more than two standard deviations later than the population average. In practice that is no testicle growth by 14 in boys, or no breast development by 13 in girls. It affects about 2 percent of adolescents, at similar rates across countries. The commonest cause by far is constitutional delay, where nothing is broken and the pulses restart late. In one cohort of 392 girls it accounted for 32 percent of cases. [24]

Why it matters — It is the same hold as in the lead story, lasting longer. It is also the group most likely to be investigated for something that resolves on its own.
20
Early breast buds, and the case for waiting

Isolated premature thelarche is breast development on its own in a young girl, without the rest of puberty following. It looks like the early stage of true precocious puberty, which does need treatment, and telling the two apart is contested. A review proposes sorting children by risk instead. Where growth rate is normal, bone age is not advanced, the pelvic scan looks prepubertal and the baseline hormone is very low, it puts active monitoring forward as the preferred approach. [25]

Why it matters — The alternative is a regular injection to switch the hormone axis off in a child who may not need it. The review is explicit that watching is an active medical process, not doing nothing.
21
Herbal medicine meta-analysis, and its limits

A meta-analysis pooled 25 studies covering 2,096 girls treated for precocious puberty with Chinese herbal medicine. It reports reductions in breast nodule diameter, uterine volume, ovarian volume and follicle volume, and finds no significant difference between the herbal treatment and the standard hormone-blocking drugs on those measures. Combining the two beat the drugs alone. The searches drew on Chinese-language databases alongside international ones, and the studies varied enough to need a sensitivity analysis. [26]

Why it matters — The outcomes measured are sizes on a scan, not final height or long-term health. A claim of equivalence to standard treatment rests entirely on that narrow set of measurements.
22
How much height is left after a first period

A review of records from 35 girls aged 12 to 16 with early or precocious puberty compared those given hormone-blocking drugs with those not given them. Twenty-seven of the 35 were treated, and the age of the first period came later in that group. Separately, a cross-sectional study of 150 Indian women over 18 found the age of a girl's first period influenced her final height, and that early menarche raised the risk of obesity. [27][28]

Why it matters — Final height is the outcome parents and doctors weigh treatment against. Both studies are small, and the second is a questionnaire rather than a follow-up.
23
The cells growth hormone actually acts on

Growth hormone is the standard treatment for children who are not growing, and it works on the growth plate, the layer of cartilage at the ends of long bones that makes them longer. Stem cells there were well described in mice, but whether humans had them was unclear. Using rare surgical specimens, researchers have now found growth-hormone-responsive stem cells in the human growth plate during puberty, matching the mouse ones. [29]

Why it matters — Responses to growth hormone vary widely between children and often fade over long treatment. Knowing which cells it acts on is the first step to knowing why.
24
Growth charts for boys with an extra X

Klinefelter syndrome, where a boy is born with an extra X chromosome, goes with tall adult height, but the childhood pattern behind it was not mapped. Researchers pooled records from six US paediatric centres covering 1,279 boys under 20 and built specific growth curves. About 20 percent of those under age 4 sat below the fifth percentile for height. A Swedish analysis of 55 boys found their pubertal growth started at 11.3 years against 11.8 in a reference group, ending at 184.6cm against 180.5cm. [30][31]

Why it matters — Without a matching chart, a doctor measures these boys against the wrong curve. The early years look like faltering growth in a group that ends up tall.
25
Turner syndrome growth, 25 years of it

Turner syndrome affects girls born with one X chromosome missing or incomplete, and short height is its commonest feature, caused by having only one working copy of the SHOX gene. Growth hormone and oestrogen replacement are the standard treatments. A 25-year study across three Taiwanese medical centres followed 107 patients treated between 1997 and 2022, comparing girls whose cells all carry the change with those where only some do, to see what predicts final adult height. [32]

Why it matters — Most published growth-hormone outcomes come from European and North American groups. A quarter-century of records from one Asian health system tests whether those results travel.
26
Male puberty is the understudied half

Between 1990 and 2016 about 1.67 times more research papers were published on female puberty than on male. Anders Juul, a children's hormone specialist at the University of Copenhagen who has studied puberty for over 30 years, says that understanding what regulates male pubertal timing may be the key to men's long-term health. Part of the reason for the gap is measurement: breast buds and a first period are visible events, while a boy's first ejaculation often goes unnoticed. [33]

Why it matters — The health links found in girls, from heart disease and diabetes to depression and early death, have no male equivalent yet. In 2023 more than half of American male deaths were classed as premature.
27
Kisspeptin's rhythm, watched live in mice

Ovulation is triggered by a surge of luteinising hormone from the pituitary, itself driven by a surge of the brain's starting hormone. Using a fibre-optic method that reads calcium activity in living animals, researchers recorded one population of kisspeptin neurons across the mouse cycle. Activity rose on a single afternoon and ran about 13 hours, made of slow waves roughly 91 minutes apart carrying fast spikes. A companion study deleted the gene in just those cells and cut fertility. [34][35]

Why it matters — The signal is a rhythm rather than a level, which is why giving the same hormone continuously shuts the system down instead of driving it. That is the basis of the drugs used to pause early puberty.
28
Three brain chemicals tracked through a lamb's puberty

Researchers followed 24 ewe lambs from four to ten months old, half born as single lambs and half as twins, measuring kisspeptin, neurokinin B and dynorphin A in their blood. All three come from the same cluster of brain cells: the first two push the puberty signal, the third holds it back. Kisspeptin and neurokinin B rose with age while dynorphin fell, and the two rises were strongly inversely correlated with the fall. Single lambs first ovulated at eight months, twins at ten. [36]

Why it matters — It is the push-and-hold pair inside one cell type, watched in an animal that can be sampled again and again. The twins were smaller at birth, which is a nutritional difference rather than a genetic one.
29
Chronic pain splits by sex at puberty

In childhood, boys and girls report chronic pain at about the same rate and process it in much the same way. That changes at puberty: girls become more sensitive to pain and develop considerably more chronic pain than boys, and the gap then holds for life. Adult women are about twice as likely as men to report widespread chronic pain. The type that rises is now called primary pain, where the pain is the problem rather than a sign of tissue damage. [37]

Why it matters — One of the largest sex differences in medicine appears at a datable moment, which makes puberty a place to look for its cause. A falling age of puberty moves when that split starts.
30
Teenage depression after the pandemic

Researchers used the ABCD study to follow the same American adolescents from before March 2020 to February 2022, across 21 sites, and sorted them into paths through the lockdown period. The design and the hypotheses were registered in advance, in March 2024. They looked for the children who had few symptoms before the pandemic and many afterwards, and for which earlier factors predicted that path: developmental stage, household, friendships, and a genetic risk score for depression. [38]

Why it matters — It separates the children who recovered from the ones who did not, in a group already being measured before anyone knew there would be a pandemic. Developmental stage is one of the factors tested.
31
Parents and children stop agreeing at puberty

Anxiety disorders start rising sharply around the beginning of puberty. Researchers examined how far a parent's report of a child's anxiety matches the child's own account, across the stretch they call peri-adolescence, when some children in a class have started puberty and others have not. The gap between the two accounts widens during that window. The team used principal components analysis to work out which parts of the two reports pull apart. [39]

Why it matters — Almost every measurement of childhood mental health leans on a parent's report. If the two accounts separate exactly when the risk climbs, the tool is weakest where it is needed most.
32
The pill, the teenage brain, and depression

Hormonal contraceptives are among the most commonly prescribed medicines for adolescent girls, and they contain synthetic versions of hormones that act on receptors spread throughout the brain. A review finds that epidemiological studies consistently report a link between use during adolescence and higher depression risk, but says a causal reading is limited, because the girls prescribed them may differ from those who are not. Brain imaging on the question is described as scarce. [40]

Why it matters — Adolescence is when depression risk in girls rises steeply on its own, which is exactly what makes the link hard to untangle. The review is a call for research, not a finding.
33
Not all teenage risk is the bad kind

A study of 127 adolescents aged 12 to 18 split risk-taking in two: negative risk, meaning potentially dangerous, and positive risk, meaning socially valued, such as speaking up or trying out for something. Working memory and planning ahead went with less negative risk-taking and had no relationship with the positive kind. One measure, effortful control, went both ways at once: less negative risk-taking and more willingness for the positive sort. The associations weakened with age. [41]

Why it matters — The standard story treats teenage risk-taking as a fault to be suppressed. If the same self-control raises one kind while lowering the other, that story is describing half of it.
34
Teenagers follow the crowd when unsure

The usual explanation for adolescents being swayed by peers is heightened reward-seeking or a need to belong. Researchers asked 166 people aged 10 to 26 to make 144 gambling decisions, some while seeing what others had chosen, under high and low uncertainty. Modelling the results as Bayesian updating, they found each person's own internal uncertainty about which option was better fell steadily with age, even though the uncertainty built into the task was identical for everyone. [42]

Why it matters — It recasts peer influence as a response to not knowing rather than a craving for approval. That points at different explanations for why the influence fades.
35
Curiosity helps adolescents more than children

Sixty young people aged 10 to 14 were shown trivia questions, rated how curious each one made them, then saw the answer and rated how interesting it was. High curiosity before the answer improved later memory for it in both children and adolescents. Interest after the answer added a further memory benefit on top of curiosity, and it did so more strongly in the adolescents than in the younger children. [43]

Why it matters — It separates wanting to know from finding something worth knowing, and only the second strengthens with age. Both effects were measured on the same material in the same session.
36
Adolescence as a second nutrition window

The Developmental Origins of Health and Disease framework holds that exposures early in life set adult disease risk, and it has concentrated on pregnancy and the first years. A paper argues that adolescence, defined as ages 10 to 19.9, belongs in it too. It reviews the evidence linking teenage diet to later heart and metabolic disease, the physiological mechanisms behind that link, and what interventions might do. [44]

Why it matters — It is an argument about where public health money should go. The window it names is the same one the lead story is about, and it is moving earlier.
37
Jumping exercise, and reading the fat signal

Sixty adolescents with body fat over 30 percent and high leptin levels were randomly assigned to a control group or to 12 weeks of plyometric exercise, which means jumping and bounding. The exercise group trained three sessions and 150 minutes a week, between July 2023 and October 2024. Twelve dropped out. Of the 48 who finished, the exercise group gained height, muscle mass and grip strength, lost fat, and showed improved leptin sensitivity and higher adiponectin. [45]

Why it matters — Leptin resistance is the state where the brain stops reading the fat signal properly, and it sits in the middle of the obesity-and-puberty chain. This is a small trial with a real control group rather than an observation.
38
A trial to test whether puberty can be slowed

LIFE-MET is a four-arm randomised placebo-controlled trial in 80 girls with overweight and early puberty. It tests metformin, a diabetes drug that improves how the body responds to insulin, with and without a lifestyle programme. Girls are assigned to metformin plus lifestyle, metformin alone, placebo plus lifestyle, or placebo alone, 20 in each arm, for six months, with a further six months of follow-up. Whether a girl gets metformin or placebo is hidden from both sides. [46]

Why it matters — Earlier work suggested that improving insulin sensitivity could slow puberty in girls with overweight, but not in a design that could rule out the placebo. This is the protocol; the results are not in.
39
Cortisol and puberty in Qom girls

Most research on the timing of puberty looks at the reproductive hormone axis. Researchers tested whether the stress axis matters too, using 777 urine samples from 46 Indigenous Qom girls aged 7 to 14 in Argentina, with the analysis registered before it was run. They asked whether higher cortisol went with an earlier first period and an earlier growth spurt. They found limited evidence for the idea. [47]

Why it matters — A pre-registered negative result on a non-Western population is rare in this field. It narrows the list of things that plausibly move the date.
40
Nerve growth factors in the puberty switch

A review examines neurotrophins, a family that includes brain-derived neurotrophic factor and is best known for keeping neurons alive, as regulators of when puberty starts. It gathers evidence that they shape wiring in the hypothalamus, change the activity of the neurons making the starting hormone, and help fold metabolic and environmental signals into that decision. Faulty signalling has been implicated in both early and delayed puberty. Human evidence is limited to genetics, blood measurements and clinical observation. [48]

Why it matters — It adds a third layer to the push-and-hold pair: the wiring itself is changing at the same time. The authors are explicit that most of the mechanism comes from animals.
41
Growth spurt and body change, matched up

The Edinburgh growth study measured healthy children born between 1972 and 1976 twice a year, recording height and the stage of visible pubertal change. Researchers modelled total pubertal growth for 157 of them, 89 boys and 68 girls, from birth length to adult height. In girls, reaching each stage of breast development later went with a smaller total growth through puberty; for one stage the correlation was minus 0.64. [50]

Why it matters — It links two things usually charted apart: how much a child grows in the spurt, and when the visible changes arrive. The data are individual curves rather than population averages.

02 Lesson why it matters

Nothing was being built. Something was being held.

The machinery for puberty is finished and working in a baby, then held shut for about nine years, so the age it arrives is set by the length of the hold and not by how long the body took to get ready.

The twist

Childhood is not the body waiting to be ready. It is the body already ready and being held, so what sets the date is how long the hold lasts.

How it works

The puberty hormone system is built and working before birth
It is switched off in early childhood, and actively held off
One gene, MKRN3, does much of the holding, and only the father's copy is switched on
Its activity falls sharply, and the pulses restart
So anything that weakens the hold moves the date earlier

The same force, elsewhere today

Where this chain is also running, in today's other stories.

Gut bacteria and the timing of puberty

the same step one layer out: a gut enzyme undoes the liver's disposal of oestrogen, so the level rises because removal stopped, not because more was made
Diet changed the brake gene's regulators

high-fat and high-carbohydrate food in rats changed the small molecules that turn Mkrn3 down, so the food did not make the animals readier, it loosened the hold
Two percent of teenagers are simply late

the same chain held longer, and nothing in it is broken; constitutional delay is the release arriving late, not machinery failing to arrive
Three brain chemicals tracked through a lamb's puberty

the holding chemical fell as the pushing ones rose, and the ovaries started working - the release is visible in the blood as a brake letting go

Where you've seen this

Seeds in winter

many will not sprout until they have had a long cold spell, so the seed is finished and the sprouting is being blocked

Sealed court records

the document is written and complete, and the date it becomes public is a rule rather than a stage of the writing

Embargoed research papers

the finding is done and checked, and the journal holds the release date, which is why the news appears everywhere at once

The catch

The hold is not the whole story. Kisspeptin is a real push and not just an absence of restraint, and in some children early puberty comes from a tumour or a hormone-making growth rather than from a hold that ran short.

And the whole of it

Nobody in this chooses the date. The child does not, the parents do not, and doctors mostly work it out afterwards from a bone-age scan. Everyone reading this went through it, and nobody could feel it happening.

03 Truth what's really going on

What is really going on

The age at which girls' breasts start developing fell about three months every decade between 1977 and 2013, and the age at which doctors call it too early, 8 in girls and 9 in boys, has not moved with it. Over roughly the same period Denmark's recorded rate of early puberty doubled in boys and tripled in girls, and South Korea's rate in girls went from 88.9 to 1,144.7 per 100,000. [3][5]

Why it works on us — A fixed age threshold feels like a fact about children's bodies, when it is a line drawn across a population that has since moved underneath it.

Who gains

Makers of the hormone-blocking drugs leuprolide and triptorelin — They are the only widely accepted treatment for progressive central precocious puberty, and the recorded rate of diagnoses doubled in Danish boys and tripled in Danish girls between 1998 and 2017. [25][5]
Sellers of Chinese herbal treatments for precocious puberty — A 25-study meta-analysis reports no significant difference from the standard drugs on breast, uterus and ovary measurements, which is a marketing claim before it is a clinical one. [26]
Growth hormone manufacturers — It is the standard treatment in Turner syndrome, and the 25-year Taiwanese study of 107 patients is about which girls respond, not whether to treat. [32]
Chemical manufacturers — The review on male fertility says the regulatory framework has not kept pace with the laboratory evidence, so the burden of proof still sits with the researchers. [20]
Researchers studying girls rather than boys — Female puberty has a visible, datable event to measure, so between 1990 and 2016 it drew about 1.67 times more published papers than male puberty did. [33]

Who pays

Girls who develop earlier than their classmates — They are more likely to be left out and bullied, and early timing tracked with more psychotic-like experiences in 11,758 children in the ABCD study. [33][6]
Boys, in what is known about their own health — The health links found in girls, from heart disease to depression, have no male equivalent yet, and in 2023 more than half of American male deaths were classed as premature. [33]
Girls whose first period comes before 13 — Pooled across 74 case-control studies, their breast cancer odds run about 15 percent higher, and about 27 percent higher if it comes before 12. [7]
Children with obesity — In two Chinese cities, 11.8 percent of girls with obesity had early puberty against 3.2 percent of girls overall, and two meta-analyses find the same direction in boys. [5][12][13]
Families of girls with early breast development alone — They sit in a grey zone the review calls contested, under repeat monitoring, while doctors decide whether it is the start of true precocious puberty or nothing. [25]
Girls treated for early puberty who are measured on height alone — The outcome the drugs are judged on is final adult height, and the study comparing treated and untreated girls followed 35 of them. [27]

What nobody knows yet

Open questions from across today’s stories — ours included.

01

How much of the rise in early-puberty diagnoses is a real change in children.

Nobody has separated a genuinely earlier population from a fixed age line meeting a shifted average, and from more children being brought in and tested. The Danish and South Korean counts are registry and insurance records, not repeat measurements of the same population. [5][3]
02

What actually lets the brake go.

The Harvard review that set out the hold-and-release picture says the mechanisms behind the restart are not completely known, and the chapter on the genetics says where MKRN3 acts inside the hypothalamus is not well understood either. [2][1]
03

Whether hormone-mimicking chemicals move puberty in real children at real doses.

The review proposing the gut-brain route pooled 87 studies and 42 of them were animal or dish work, and it argues direct hormone disruption does not explain the effect on its own. [16]
04

Whether treating early puberty with hormone-blocking drugs helps anything beyond final height.

The premature thelarche review frames the whole question as contested, and proposes sorting children by risk precisely because the case for treating is not settled. The study comparing treated with untreated girls on final height followed 35 of them. [25][27]
05

What sets the timing of male puberty.

About 1.67 times more papers were published on female than male puberty between 1990 and 2016, and a boy's first ejaculation, the closest equivalent to a first period, often goes unnoticed, so there is no easy event to date. [33]
06

Whether hormonal contraceptives in adolescence raise depression risk.

The link turns up consistently in population studies, but the girls prescribed them may differ from those who are not, and the brain imaging on the question is described as scarce. [40]
07

Whether Chinese herbal medicine does anything for precocious puberty.

The meta-analysis pooled 25 studies and 2,096 girls and reports equivalence to standard drugs, but only on scan measurements of breast, uterus and ovary size, and the studies varied enough to need a sensitivity analysis. [26]
08

Whether the brain's adolescent phase really runs to about 32.

It rests on one study of nearly 4,000 scans, and the researcher who led it says it describes the pattern of change rather than how people in their late twenties behave. [9]
09

Why urban boys in the Chinese cohort had early puberty at 5.7 percent and suburban boys at 0.7 percent.

The gap inside one country is wider than the gap between weight groups, and the paper reports it without an explanation. [5]

04 Hope carry this

Stem cells in the human growth plate, the cartilage at the ends of long bones that makes children taller, were only clearly described in mice. Using rare surgical specimens, researchers have now found them in the human growth plate during puberty, and they respond to growth hormone.

Also true today

For the first time, the brain cells that release the signal starting puberty have been grown from stem cells in a dish, and they secreted kisspeptin into the culture.
Forty-eight adolescents with high body fat finished twelve weeks of jumping exercise, three sessions a week. They gained height, muscle and grip strength, lost fat, and their bodies read the fat-signalling hormone leptin better than before.
For a young girl with early breast development but normal growth, unadvanced bone age, a prepubertal scan and a very low baseline hormone, a review of the evidence now puts careful watching ahead of switching her hormone axis off.