How hospital birth times reshape the birth chart

Key findings

• In all three countries most babies born in 2024 had a day chart, the Sun above the horizon: 63.6% in Brazil, 67.5% in Japan and 59.3% in the USA, against 50% if births were spread evenly over the day. Brazil rose from 59.8% in 2006 and Japan from 63.4% in 2015, while the USA slipped from 60.7% in 2016.
• Not caesareans alone. Japan has the highest share of day charts with about one caesarean in five births. In Brazil about two thirds of the rise since 2006 (+2.6 of +3.9 points) comes from the larger caesarean share; the rest is the net of caesarean births moving further into the day (+2.2) and vaginal births moving slightly out of it (−0.9).
• For the Sun, each Placidus house is one sixth of its own day or night, so the Sun’s house works as a clock. The commonest house for the Sun is the 10th in Brazil (1.53 times its share with even birth times) and the 8th in Japan (1.77 times); in the USA the 9th to the 11th are nearly tied (×1.24–×1.27). Mercury and Venus follow the Sun.
• Over a whole year the hour of birth barely changes how common each Ascendant sign or Moon sign is (at most 0.28 and 0.12 percentage points in any country-year), but it strongly reshapes combinations tied to the date: in Brazil some Sun–Ascendant pairs are 0.53 times and others 1.55 times as common as with even birth times.
• The expected share of Mars in Michel Gauquelin’s key sectors for babies born in a single year is anywhere from 12.4% to 22.4% in Japan (14.8–19.7% in Brazil), not a fixed one sixth (16.7%). Pooled over many years, modern birth timing still adds 0.8–1.0 percentage points, for a structural reason: Mars spends 53% of the time within 60° of the Sun and inherits the daytime peak of births.
• A column in the Brazilian newspaper O Tempo estimated that about 80% of people born in Brazil today have at least the Sun, Mercury and Venus in the upper half of the chart. Measured: 55.7% in 2024, the highest of the 19 years (46.8–55.7%).

Day charts: the Sun above the horizon at birth

Each result is shown on a ladder of four comparisons, built for the same places, so every step isolates one factor. Astronomy: the same number of births on every date and at every hour. Real dates: the real number of births on each date (seasons, weekends, holidays; for the USA month and weekday only), still evenly over the hours. Non-induced hours: the real dates spread over the day like births that were not induced (Brazil and USA: vaginal births without induction; Japan: births at midwife homes and at home). Real: every birth at its recorded time.

The real calendar moves the share by less than a third of a percentage point; the hours do the rest. The non-induced column is not “natural” birth: in Brazil vaginal births without induction still peak between 8 and 11 in the morning and give 52.6–53.2% day charts (2012–2024; the induction field is incomplete before 2012).

Births at Japanese midwife homes and at home keep the old night-time peak and give fewer than half day charts (49.0%). They are a small, particular group, so the Japanese column is not comparable with the other two.

The US files also record where a birth took place. Births planned at home or in a freestanding birth center (542,277 with a recorded time in 2016–2024; 1.3% of births in 2016 and 2.0% in 2024) are most frequent between 1 and 5 in the morning. 37.2% of them fall between 8 a.m. and 6 p.m., against 51.8% of all US births, and 45.9% have a day chart. These are their own recorded times, not a comparison column; the files give them no place, so they carry the county shares of all births.

When babies are born

Not caesareans alone

In Brazil caesareans rose from 45% of births in 2006 to 61% in 2024, and day charts from 59.8% to 63.6%. The share of caesarean babies born with the Sun above the horizon rose from 67.3% to 71.4%, while that of vaginal births slipped from 53.6% to 51.7%. Splitting the +3.9 percentage points: +2.6 come from more caesareans, +2.2 from caesareans moving further into the day and −0.9 from vaginal births moving slightly out of it. This is an accounting split with midpoint weights; holding the 2006 or the 2024 day shares fixed instead gives the caesarean share 55% or 79% of the rise instead of 67%. Across the 513 state-years of the 27 Brazilian states the day-chart share and the caesarean share are correlated at 0.76 — a correlation, not a measured effect of caesareans.

Japan has the highest share of day charts of the three with about one caesarean in five births (21.6% of deliveries in hospitals and clinics in 2020, from the Medical Facility Survey, as cited by the Ministry of Health, Labour and Welfare↗). Its births peak between 1 and 3 in the afternoon, which points to the scheduling of care in general, not only the operation, setting the clock.

In the USA caesareans stayed at about a third (31.9% in 2016, 32.3% in 2024) and labour was induced in 24.5% of births in 2016 and 33.9% in 2024, yet day charts fell slightly, from 60.7% to 59.3%. The share of births between 8 a.m. and 6 p.m. fell among induced births (50.4% to 47.2%) and among vaginal births without induction (45.8% to 44.6%), while caesareans stayed near 63%. We report this without attributing a cause.

The Sun’s house is a clock

Placidus cusps are found by dividing into three equal parts of time the arc that each point of the zodiac travels above the horizon, and the arc below it. For the Sun this means its house is one of six equal slices of its own day or of its own night — pairs of seasonal (unequal) hours in local solar time, not clock time: the 12th, 11th and 10th are the morning up to solar noon, the 9th, 8th and 7th the afternoon, the 6th to the 4th the evening until solar midnight and the 3rd to the 1st the small hours. Whatever concentrates births at certain hours concentrates the Sun in certain houses. Ratios compare real births with the same dates at even times, all years pooled.

In Brazil the Moon’s house hardly moves (every house within 0.2% of its even-time share, all years pooled): the Moon circles the sky every month, so the hours average out. Mercury (×0.55–×1.43) and Venus (×0.61–×1.33) never stray far from the Sun and follow it.

What birth times do not change, and what they do

Over a full year the sky at any fixed clock hour turns through all twelve rising signs, so crowding births into certain hours hardly changes how common each Ascendant sign is: the real share differs from the even-time share by at most 0.28 percentage points in any country-year. The Moon sign moves even less (at most 0.12 points). The Sun sign depends on the date only: seasons move it by up to about one percentage point, hours not at all.

Anything tied to the date is different. Given the Sun sign, the birth hour decides the Ascendant, so the Sun–Ascendant pairs are reshaped strongly. The table shows real Brazilian births per 10,000 (all 19 years), coloured by the ratio to the same dates at even hours (×0.53–×1.55; Japan ×0.60–×1.88). Adding the Moon sign to make the “Big Three” (Sun, Moon and Ascendant signs) changes almost nothing about this: the hour effect on the Big Three is the Sun–Ascendant effect.

▲ darker gold: at least 1.3 times as common as with even birth times; ▼ darker blue: 0.7 times or less; lighter tones: 1.1 to 1.3 and 0.7 to 0.9. With the same dates at even hours the pairs would range from 52 to 84 per 10,000 (not a flat 69, because signs rise in unequal times); real Brazilian births range from 28 to 125. Japan and the USA are in the CSV.

Mars and the Gauquelin sectors

Michel Gauquelin divided the daily circle of each planet into 36 sectors and reported that Mars sat more often than expected in the “key” sectors just after rising and just after culmination (sectors 1–3 and 10–12) in the charts of sports champions. Much of the debate turned on what “expected” is. For babies born in a single year the answer swings with the planet’s position relative to the Sun, which repeats about every 2.1 years:

Part of the yearly swing is astronomy alone (the first column). The rest is birth timing, and it does not cancel out: pooled over all years real births put Mars in the key sectors 0.8–1.0 percentage points more often than the same dates at even times. The reason is that Mars is not spread evenly around the Sun: it spends 52.8% of the time within 60° of it (an even spread would give 33.3%; Jupiter 38.7%, Saturn 36.3%), so it inherits a blurred copy of the two daytime peaks of births, late in the morning and in the afternoon, which lie either side of sectors 10–12. A model built only from that (the Sun’s sector distribution shifted by where Mars stands relative to the Sun) follows the year-by-year results closely (correlation 0.99 in Brazil) and, if today’s Brazilian hours applied throughout 1900–2099, predicts +1.02 points.

This does not test the Mars effect. It says what a fair control group looks like today: born in the same country and years, at the same hours, not a fixed one sixth of the circle.

Prior work

• The idea of an expected frequency built from the real daily rhythm of births is Michel Gauquelin’s, from 1955; controls born at the same times and places were proposed by Marvin Zelen (1976), and the French committee CFEPP built its expectation by randomly recombining real birth records (Benski et al., 1996). See the summary by Skepsis↗ and Geoffrey Dean, “Is the Mars effect a social effect?” (2002)↗.
• That caesarean scheduling fills the upper half of the chart, favours day charts and thins out some Sun–Ascendant combinations was raised by the astrologer Wendy Stacey from about 90,000 births at four Bristol hospitals (The Astrology Podcast, episode 214↗).
• The hourly rhythm of births is well documented: the NCHS data brief for the USA in 2013↗, England 2005–2014 (PLOS ONE, 2018)↗ and Japan (J Epidemiol, 2002)↗.
• What this study adds: complete national registers rather than a hospital sample, the chart computed birth by birth for Brazil and from hour-level national counts for Japan and the USA, three countries, a 19-year series, the ladder that separates astronomy, calendar and hours, and open code and data.

Checking a published estimate

A column in the Brazilian newspaper O Tempo on 24 September 2026 (in Portuguese↗) estimated, and called it an estimate, that about 80% of people born in Brazil today have at least the Sun, Mercury and Venus in the upper half of the chart (houses 7–12). Measured: 55.7% of babies born in 2024, the highest of the 19 years (46.8–55.7%, depending on how far Mercury and Venus stand from the Sun); the Sun alone 63.6%. The estimate counted every caesarean as a daytime birth, and the column itself notes that urgent caesareans happen at any hour; measured, 71.4% of caesarean babies are born with the Sun above the horizon.

Method

• Brazil: SINASC, the national register of live births (Ministry of Health, DATASUS), 2006–2024; files before 2006 carry no birth time. Every birth with a valid time (99.63–99.95% a year) is placed at the seat of its municipality of birth and converted from the local clock to UT with that place’s time zone and daylight-saving history.
• Japan: vital statistics, e-Stat table 0003411915 (births by month, day and hour), 2015–2024, national only; spread over the 47 prefectures by their share of births and placed at the prefectural offices. Births within an hour are spread evenly over it.
• USA: NCHS natality public-use files, 2016–2024. They give the month, the weekday and the clock time but no day of the month and no place. Births of each month and weekday are spread over the matching dates; places come from Census Bureau county birth counts. To measure the error of exactly this, we removed the same information from Brazil’s 2018 data (a year with daylight saving time) and repeated the US method: day charts moved by 0.11 points, the Sun’s house shares by up to 0.45, Mars’s by up to 0.10, and Sun–Ascendant pairs by up to 8% of their value (5% for 2024).
• The chart: Swiss Ephemeris positions; day chart means the centre of the Sun geometrically above the horizon, which is the same as the Sun in Placidus houses 7–12; Placidus houses by ecliptic longitude; whole-sign houses; Gauquelin sectors from the planets’ true positions; tropical signs (Lahiri sidereal signs are in the CSV). The formulas were checked against Swiss Ephemeris on 2,000 random charts with 0 mismatches, and the main figures were recomputed independently with a second astronomy library.

Limits

• These are babies born in these countries and years. Nothing here describes adults born before the data start, whose births most likely followed a more nocturnal rhythm (as births at Japanese midwife homes and US births planned out of hospital still do), or people born elsewhere.
• Recorded times are what the hospital wrote down. Rounding is small and fading: in Brazil 8.1% of times ended in :00 in 2006 and 2.9% in 2024 (1.7% expected); in the USA about 1.8%.
• The US and Japanese figures rest on the national hour profile applying everywhere in the country; Japan also on 2023 prefecture shares. US births planned out of hospital are placed with the county shares of all births, which may not match where they happen. Brazilian places are municipality seats.
• A correlation between caesareans and day charts is not a measured effect of caesareans, and none of this says anything about whether astrology works.