The mundane timeline of a month मेदिनी ज्योतिष

Pick any month. Horā computes, from that month's real sky, the mundane conditions that were in force across it — and prints beside each one what the record holds for the past periods in which the same condition held. Counts, with their sample sizes, against a disclosed baseline.

Every number on this page is about something that has already happened. This page does not say that anything is going to happen. Horā issues no forecasts about wars, disasters or markets, and a condition being in force in a month says nothing whatever about that month.

← September 1931 →

Month Year

The record holds no event inside this month

September 1931 lies inside the tested period, but no event of this curated set is dated inside it. That is the ordinary case: the chosen events fall in 118 of the 1,512 months of 1900–2025.

An empty list means the record is silent about this month — not that the month was quiet. What happened in the world that month and what entered this small, curated set are not the same thing.

4tested conditions in force during this month

30 / 30days with at least one of them open

0recorded events dated inside this month

This month's sky, day by day

Each band shows the days of the month on which that condition held. Days are read at 12:00 UTC. The bands carry no favourable or adverse colour, deliberately: events falling more often inside a window is not good news.

Days of the month

Tested condition

Within 18 days either side of an eclipse

1–30 September 1931

In force on 30 of the month's 30 days.

What already happened while this same condition held — counted across the whole record:

Series	On event days	Across all sampled days	Reading
geophysical events	13 / 46 (28.3%)	30.6% / 46,021	near the baseline — no signal
political events	14 / 44 (31.8%)	30.6% / 46,021	near the baseline — no signal
market events	10 / 45 (22.2%)	30.6% / 46,021	near the baseline — no signal

Tested against: geophysical events, political events, market events. Categories this condition was not tested against are left out rather than shown as a zero; that restriction is this platform's stated configuration, not a statement of the texts.

Classical basis. Proximity to an eclipse season — the luminaries near the nodal axis. The eighteen-day span is Horā’s stated convention.

Tested condition

Budha retrograde or combust

1–13 September 1931

In force on 13 of the month's 30 days.

What already happened while this same condition held — counted across the whole record:

Series	On event days	Across all sampled days	Reading
market events	22 / 45 (48.9%)	43.9% / 46,021	near the baseline — no signal

Tested against: market events. Categories this condition was not tested against are left out rather than shown as a zero; that restriction is this platform's stated configuration, not a statement of the texts.

Classical basis. The price (argha) and motion chapters of the Bṛhat Saṁhitā link anomalous motion of the significator of trade to disturbance in prices. Retrogression and a 12° combustion orb are Horā’s stated operationalisation.

Tested condition

Graha-yuddha — two tārā-grahas within 1°

5 September 1931

In force on 1 of the month's 30 days.

What already happened while this same condition held — counted across the whole record:

Series	On event days	Across all sampled days	Reading
geophysical events	5 / 46 (10.9%)	8.0% / 46,021	above the baseline — at the 5-hit display floor
political events	3 / 44 (6.8%)	8.0% / 46,021	too few to read
market events	6 / 45 (13.3%)	8.0% / 46,021	above the baseline

Classical basis. Bṛhat Saṁhitā, the graha-yuddha chapter: a war between planets is said to distress nations, and the people of the defeated planet to suffer.

Ingress

Maṅgala enters Tulā

7 September 1931

7 September 1931 — first found in the new rāśi at 12:00 UTC on this day.

Classical basis. An ingress is the crossing of a sidereal (Lāhirī) rāśi boundary, sampled at 12:00 UTC each day. Only Maṅgala, Guru, Śani and Rāhu are shown; Budha and Śukra change sign every few weeks and no tested condition is keyed to their ingress.

Tested condition

Within 14 days after an eclipse

12–25 September 1931; 27–30 September 1931

In force on 18 of the month's 30 days.

What already happened while this same condition held — counted across the whole record:

Series	On event days	Across all sampled days	Reading
geophysical events	10 / 46 (21.7%)	17.4% / 46,021	above the baseline
political events	7 / 44 (15.9%)	17.4% / 46,021	near the baseline — no signal
market events	5 / 45 (11.1%)	17.4% / 46,021	near the baseline — no signal — at the 5-hit display floor

Classical basis. Bṛhat Saṁhitā, the eclipse chapters: effects are said to follow an eclipse. The fourteen-day window is Horā’s stated convention, not the text’s number — the text gives no count of days.

Eclipse

Solar eclipse

12 September 1931

12 September 1931, maximum at 04:41 UTC.

Not tested by the record None of the seven tested conditions is keyed to this event on its own, so the record offers no count for it. It is listed because it is a fact of this month's sky, and no number is attached to it. The tested window an eclipse opens appears above as its own band, and the counts are given there.

Classical basis. The eclipse maximum comes from the same astronomical routine this platform's eclipse catalogue was built with. The time is UTC.

Station

Budha turns direct

14 September 1931

14 September 1931 — the change of direction is first seen at 12:00 UTC on this day.

Classical basis. A station is read from the change of sign of the body's speed, sampled at 12:00 UTC each day. No time finer than the day is computed.

Station

Śani turns direct

22 September 1931

22 September 1931 — the change of direction is first seen at 12:00 UTC on this day.

Classical basis. A station is read from the change of sign of the body's speed, sampled at 12:00 UTC each day. No time finer than the day is computed.

Eclipse

Lunar eclipse

26 September 1931

26 September 1931, maximum at 19:48 UTC.

Classical basis. The eclipse maximum comes from the same astronomical routine this platform's eclipse catalogue was built with. The time is UTC.

What the record holds for this month itself

These are the events in the curated set whose date falls inside this month, with the conditions that held at each event's own recorded moment. It is a record, not a cause: nothing below says that a planetary configuration made any of it happen.

The record holds no event dated inside this month.

What this page stands on

The events135 dated events across three categories. 5 of them pre-date 1900: they are counted in the sample sizes beside every condition, but the eclipse catalogue begins in 1900, so for the two eclipse-window conditions those 5 can never register as hits — they widen the sample and can never raise the count. They also fall outside the months this selector can open.
The months118 of the 1,512 months of 1900–2025 hold at least one recorded event. For the rest, the record is silent.
The eclipsesThis month's eclipses come from the same catalogue the correlation arithmetic ran on, so this page and the evidence page cannot disagree about a past month.
The daysEvery day is read at 12:00 UTC. Events are tested at their recorded time where the record has one and at 12:00 UTC where it does not — so a band edge and an event can differ by a few hours.

Months the record holds more than one event in

These are not recommended months — they are simply the months in which more than one event of this curated set falls. The figure in brackets is how many.

October 1929 (3) December 1971 (2) October 1987 (2) October 1997 (2) September 2001 (2) January 2008 (2) May 2008 (2) September 2008 (2) March 2020 (4)

How this timeline is built

The conditions are the same sevenThese are the seven conditions of the same tested set the whole record was measured against. No new condition has been added here, and no new rule invented.
The counts come from the same tableEvery count on a row is read out of the engine's own table — the one the evidence page prints — rather than recomputed. The two pages therefore cannot disagree about a row.
Windows and orbs are stated conventionsThe texts say effects follow an eclipse; they do not say fourteen days. Every window and every orb is this platform's decision, and it is printed with the row it belongs to.
Fewer than five hits means no directionA row with fewer than 5 events inside the window is reported as too few to read rather than as above or below. That floor is this platform's display convention, not a classical or statistical standard.

What this page will never do

The fence the geo-political record carries applies here unchanged, with two lines added for what a month view makes newly tempting.

No forecast of any index, price, currency or commodity, on any horizon
No buy, sell, hold or timing suggestion, direct or implied
No claim that a planetary configuration caused, or will cause, anything
No “favourable period for investment”, and no auspicious dates for a trade
No personalised financial, political or legal advice — Horā is not a licensed adviser and will not act like one
No month called good or bad for anything, and no band turned into a reason for caution or for confidence
No world event attached to any living person's chart

What this page offers is historical correspondence, counted after the event, with its limitations stated. Correspondence is not causation, a curated set of famous events is not a sample of history, and none of it is a basis for a financial, political or personal decision. Treat any service that turns a planetary configuration into a trade or a warning with the suspicion it deserves.

What this timeline cannot tell you

A month's bands and the record's counts are not the same thing.The bands describe this month's sky. The counts belong to the whole record. Printing them on one row does not mean one follows from the other.
The events were chosen, not sampled.No event entered the set because the sky was interesting, but many real events never entered it because they never became famous. That is selection bias, and it cannot be argued away.
Every category holds fewer than fifty events.One event moving in or out of a window shifts a ratio visibly. That is why counts are printed here rather than percentages.
Correspondence is not causation.Even a row above the baseline would show only that events and a condition co-occurred more often than the all-days baseline rate, on this particular set.
Most rows found nothing.That is the honest headline, and those rows are here at full size.
Nothing here is a forecast.This page reports what already happened. A condition being in force in a month says nothing whatever about that month.

Back to the geo-political record Read the full test results