The evidence, layer by layer
One established fact sits at the bottom; everything above it is held at a stated confidence. The cards below separate what is documented from what is finding, caveat, or rival reading.
The Somerton Man is Carl Webb
2022 DNA identification via rootless-hair genome data and a 4,000-person family tree (Abbott & Fitzpatrick). Formal coronial confirmation outstanding.
Webb was searching for a woman
The documented thread points to Glenelg: her number on the book, his death 400 m from her door. The rival Dorothy-in-Bute claim has no 1948 record behind it and a 1949 Melbourne electoral roll against it.
A paternity belief drove the search
DNA excludes Webb as Robin Thomson's father. The motive survives only as something Webb may have believed — never as fact.
The code is a travel record
Consistent with the evidence, modestly supported by frequency analysis, untested by sequence-and-connectivity. The Map Lab shows both the support and the warning.
The body and the scene
Found 6:30 am, 1 December 1948, against the seawall at Somerton, opposite the Crippled Children's Home, legs extended, half-smoked cigarette on the collar. Seen motionless the previous evening.
Autopsy: acute gastric haemorrhage, spleen about three times normal size, last meal a pasty 3–4 hours before death. Poison suspected; none identified.
The pockets
An unused second-class rail ticket for the 10:50 am to Henley Beach. A used bus ticket toward Glenelg from North Terrace. Juicy Fruit, an American comb, Army Club packet holding Kensitas cigarettes. No wallet, no papers; clothing labels removed.
The suitcase
Checked at Adelaide Railway Station 11:00–11:15 am, 30 November; recovered 14 January 1949. Tied to the body by orange waxed thread matching a trouser repair. Garments marked T. Keane — the name of Webb's brother-in-law.
The slip and the book
"Tamám Shud" — ended — rolled in a fob pocket, torn from a Rubáiyát that surfaced from a car parked near Glenelg. On the inside back cover: the pencilled letters, a telephone number, and by some accounts a second number never identified.
The number and the address
X3239 reached Jessica Thomson at 90A Moseley Street, Glenelg — about 400 m from the body. The number was semi-public: Prosper Thomson's car classifieds printed it from July 1947, the one paper-trailed route by which Webb could have held it.
52 Moseley Street, given in some secondary sources, is wrong; 90A is the 1948 electoral-roll address.
The code is a copy of a copy
Faint pencil indentations, photographed and reportedly over-written. First letters of lines 1, 3 and 4 are disputed between W and M. Any reading that hangs on a single letter hangs on a seventy-five-year-old rubbing.
Initials, not cipher
1949 naval cryptographers: letter frequencies sit closer to the initial letters of English words than to any cipher table. University of Adelaide work later eliminated 20+ cipher types and reached the same conclusion. The question becomes: initials of what?
The test that cannot fail
All 17 distinct letters have at least one Adelaide-reachable candidate. With 575 verified stations plus tram and bus stops, set-membership matching always succeeds — for the code, or for any random string. RPFA exists to replace that broken test with a constrained one.
Horse-betting initials
Webb bet on horses; Abbott favours a betting-list reading; and it benefits from the same initialism finding. Co-equal with the itinerary reading on current evidence. The itinerary's edge: the coastal G–A–B ending, and a searcher's obvious reason to keep a record.
Espionage and suicide
The spy reading is badly weakened by the Webb identification: a Melbourne instrument maker with a failed marriage and no intelligence trace. The suicide reading is strong and compatible — the letters concern how he spent his last days, not how they ended.
Map Lab
Two instruments on one 1948 map. The RPFA lens tests where a compact search base could sit: drag the circle, set its radius, and watch which code letters it can satisfy from plate-verified locations. The route tracer walks the five lines of the published itinerary reading.
Lens radius
Drag the coral circle on the map. The dashed ring shows the smallest radius that would cover all 17 letters from the current centre.
Letter coverage
What this shows — and what it cannot
At small radii almost nowhere qualifies; full coverage opens up only in a narrow inner north-western band, with Q the binding constraint. That bounds where a compact search base could sit. It does not prove the letters are places: that requires the sequence-and-connectivity test, which this lens deliberately does not perform.
The case in time
Documented events with their record types. The hatched block is the nineteen-month documentary void; the itinerary hypothesis treats the code as potentially recording movements inside it.
The final day, hour by hour
Solid flags are documented; hatched bands are gaps or inferences.
Sources & data
The primary cartographic source is the 1938 South Australian Railways Commissioner's map, PP No. 47 (Schedule of Railways, 30 June 1938, signed R. H. Chapman, Chief Engineer for Railways). Every gazetteer location cites a plate below. Drag scans in to attach them; they persist in this browser and each station's plate becomes viewable in place.