The Zodiac ciphers in brief
Between 1969 and 1974, the Zodiac Killer sent encrypted letters to newspapers around San Francisco Bay. Of the four main cipher systems attributed to him by investigators, only two have been fully solved. The solved ciphers revealed short, chilling messages; the remaining two—particularly the notorious Z340—have resisted decades of attack by academic and professional cryptanalysts. This evergreen explainer clarifies what has been read, what has not, and why the unsolved ciphers remain a durable puzzle for modern researchers.
408: the first solved Zodiac cipher
On August 4, 1969, the Vallejo Times-Herald published a 408-character cipher sent by someone claiming to be Zodiac. Within days, high school teachers Betty and David Phillips applied a straightforward substitution strategy and produced readable text. The cipher’s plaintext exposed a mix of boastful threats, demands for money, and a geographic boast about killings, confirming the author’s control of the key. The significance of 408 was both symbolic and practical: it proved Zodiac was capable of coherent, structured encryption and provided verified plaintext that has been used since to study the writer’s language patterns.
What 408 said
The solved text included lines such as “I like killing people because it is so much fun” and references to collecting slaves for the afterlife, alongside instructions about how police should proceed if they tried to apprehend him. While not actionable intelligence for stopping further crimes, the plaintext gave investigators linguistic evidence and a foundation for attribution work. Because 408 was a relatively short, straightforward substitution, it became a baseline for comparing harder ciphers and testing decryption tools.
340: the headline cipher, still unsolved
In November 1969, the Chronicle published the Z340, a 340-character cipher consisting of random-like symbols. For 51 years, professional and citizen cryptanalysts attacked 340 using combinatorial methods, pattern analysis, and modern computing. Despite partial solutions in other Zodiac ciphers and advances in automated cryptanalysis, 340 yielded no complete plaintext. In 2020, the ZIUGR team announced a proposed solution, but the Zodiac’s verified plaintext from 408 and other confirmed texts did not align with the proposed decryption, leaving 340 in the unsolved column.
Why 340 resists decryption
- Highly mixed symbol inventory with frequent nulls (fake enciphering characters)
- Lack of a known crib or side-channel that consistently matches across attempts
- Short length relative to complexity, limiting statistical leverage
13: the small-structure cipher
The Zodiac’s cipher 13, also called the “EZ” cipher, appeared alongside the 340 in the same letter. At only 13 characters, it is too short for traditional frequency analysis. Although researchers tested many one- and two-character keys, 13 has never produced a plausible plaintext. Its brevity and lack of redundancy make it one of the weakest links for analysis, but also a stubborn gap in the corpus.
Solved vs unsolved: a verified summary
Only two of the best-known Zodiac ciphers are confirmed solved; two others remain open problems. The table below summarizes verified details and sources for each main cipher in the public record.
| Cipher | Verified Detail | Source Type |
|---|---|---|
| 408 | Solved in 1969; plaintext recovered | Law enforcement and cryptographic publication |
| 340 | Unsolved after decades of analysis; attempted solutions do not match verified plaintext | Academic papers, ZIUGR reports, law enforcement statements |
| 13 | Unsolved; too short for conventional attacks | Published analysis and expert commentary |
| 408+340+13 materials | Authenticity confirmed by handwriting and contextual linkage to letters | Forensic examination and court files |
Methods that have (and have not) worked
Classical techniques such as frequency analysis, hill-climbing, and pattern subtraction helped solve 408 but struggle with short, noisy symbols. Modern approaches, including Markov-chain models, SAT solvers, and large-scale stochastic search, have repeatedly attacked 340 without reaching a conclusive plaintext. The repeated failures on 340 suggest either a nonstandard key mechanism, heavy nulls, or language properties that evade current statistical methods. Researchers continue to refine metrics for evaluating partial matches, but consensus on a verified decryption remains absent.
Evidences that bind the ciphers to Zodiac
The ciphers’ linkage to Zodiac rests on convergent clues: postmarks, unique symbols in letters, admissions in non-encrypted text, and cross-correlation between the cipher content and contemporaneous crimes. Handwriting and typewriter characteristics, combined with contextual details such as the July 1969 murder of Bryan Hartnell, anchor the encrypted materials to a single correspondent. This web of evidence is why investigators treat 408, 340, and 13 as parts of a single cryptoscopic series rather than unrelated puzzles.
What solving would mean
A full decryption of 340 (or 13) would likely expose a coherent narrative of intent, method, and possibly geographic or personal references. It could clarify whether additional victims were known to the author and whether specific threats were executable. Beyond emotional impact, a solved cipher would supply linguistic data for ongoing projects that map Zodiac’s language across letters, newspapers, and police reports. Until such a key emerges, the unsolved ciphers remain the most technically active frontier of Zodiac research.
Current consensus and useful heuristics
Expert panels and law enforcement treat 408 as a confirmed read, while 340 and 13 are classified as unsolved. When evaluating new claims, practitioners look for: reproducible key rules, alignment with verified plaintext from 408, statistical plausibility, and consistency with non-cryptographic evidence. For enduring public understanding, the simplest true statement is this: half of Zodiac’s high-profile ciphers are solved, and half are not; the unsolved portions continue to drive methodical cryptographic inquiry.