Zodiac killer info

The Zodiac Killer (a.k.a "the Zodiac" or just "Zodiac") was the pseudonym of an unidentified serial killer who operated in Northern California in the late 1960s. The Zodiac murdered five known victims in the San Francisco Bay Area between December 1968 and October 1969, in Benicia, Vallejo, Napa County, and San Francisco. The Zodiac claimed to have murdered 37 victims, and he has been linked to several other cold cases.


The goal of this website is to present information about the Zodiac killer case in an unbiased way. Each piece of information is accompanied by a link to the primary document(s) that supports it. Information that is missing sources or that is not clear will be indicated in red text. If you want to contribute information to this website, or if you want to suggest changes, please contact the site administrators.


Police sketch of a suspect seen at the Presidio Heights murder of Paul Stine. The sketch was done by SFPD police artist Juan Morales, made with the help of the Robbins kids, who witnessed the murder. The amended drawing at right was produced after a one-on-one session with Rebecca Robbins. SOURCE?

Police identi-kit composite of a suspect seen at Lake Berryessa by 3 female college students at the lake that day. 

The "official" police sketch was apparently a "corrected" version derived from the identikit composite.

Drawing of the Zodiac at Lake Berryessa based on the witness testimony of Bryan Hartnell. Sketch by Juan Morales of SFPD. SOURCE?

A NOTE on time estimates

The time estimates in this website come primarily from witness statements. It is important to be aware that any time estimate may be (and probably is) inaccurate by several minutes. In my opinion, very few time estimates in a case like this are likely to be accurate within even 1 or 2 minutes. 

The accuracy of any time estimate is affected by numerous factors: 1. how recently did the witness look at a clock (or otherwise know the time), either before or after an incident took place; 2. how good is the witness at estimating times or time durations; 3. how good was their memory; 4. how recently did the event occur, etc. As memory fades, the accuracy of time estimates decreases. Additionally, it is likely that many witnesses "round" time estimates to the nearest 5 or 15 minutes on the hour. For example, a witness would be more likely to estimate that an event happened at "11 PM" or "11:15 PM," instead of a more precise time like "11:08 PM." 

Because of the above reasons, we should assume that any time given in a witness statement may be off by as much as 5, 10, or even 15 minutes... or more. For this reason, any precise reconstruction based on time estimates must be considered conjectural only. 

It is probably correct to assume that some time estimates are more accurate than others, and this depends on several factors. For example, Nancy Stover's 12:10 AM time estimate for the phone call reporting the Blue Rock Springs murder is probably more accurate than normal, as it was probably part of her job to log the times at which calls were received (Is it possible to verify this?). This 12:10 time might in fact be precise to the minute, or within a few minutes... but this is the exception, not the norm. MOST witnesses do not know the time when events occur with any degree of accuracy. 

In this website, in some cases, I have based time estimates more on logic and reason rather than strictly on witness statements. The Blue Rock Springs timeline, for example, is based in part on working backwards from Nancy Stover's 12:10 time estimate, using the duration of events (for example the fact that it is a 7-8 minute drive from the park to the house where the teens placed the call to Nancy Stover.) This method may be more accurate than relying solely and strictly on the time estimates given by witnesses such as Mike Mageau or Richard Hoffman, that are, in all likelihood, less precise. But even here, the timeline is mostly conjectural.

A NOTE on witness testimony

Witness testimony, as reported in police reports, and in other sources, is undoubtedly one of the most important primary sources in criminal cases. However, witness statements should not be taken as being 100% accurate. Eyewitness testimony has been shown to be notoriously inaccurate: witnesses may misremember things, and they might simply not notice things in the first place. When reading police reports it is important to keep in mind that much of what is written may have been misremembered, not noticed, or simply guessed at. 

It is also important to note that witness statements are not usually verbatim transcripts of what was said by a witness, but rather are summarized responses to questions asked by investigators. 

For example, much has been made out of the fact that James Owen changed his estimate of the distance between David Faraday's car and the car parked next to it at the turnout on Lake Herman Road. In his first police statement, Owen said the cars were "about 10 feet apart," but then in a second statement this was changed to "three or four feet." What happened here? In all likelihood, there is an simple explanation for the discrepancy. It is unlikely that Owen noticed with any degree of accuracy how far apart the cars were, and was just guessing. And he was likely responding to a question asked to him. The reporting police officer probably asked, "how far apart were the cars?" to which Owen may have answered something like, "I don't know. Maybe ten feet, I'm not sure." The officer would summarize this by writing that the cars were "about 10 feet apart apart." Owen may have later reconsidered this, and given a slightly different answer. People often do not notice such things with any degree of accuracy. This should be kept in mind when reading witness statements.