The Lab That Failed Twice: MSP Forensic Science and the Michigan Murders Archive
By Rita Williams | Clutch Justice | July 12, 2026
- Article 01: The Foundation Was Junk
- Article 02: The Uninvestigated Six
- Article 03: From These Cases to the Federal Bench
- Article 04: The Lab That Failed Twice (current)
The Michigan State Police Crime Laboratory produced two distinct institutional failures in the Michigan Murders archive. The first: it supplied the forensic testimony that supported the Collins conviction in 1970, relying on methods that have since been formally repudiated. The second: it contaminated or mishandled the physical evidence in the Jane Mixer case, generating a DNA result in 2002 that fractured the Collins attribution and produced a separate conviction. The lab is not a passive repository in this record. It is an active institutional actor whose decisions shaped what the archive contains and what it does not.
What the Lab Supplied in 1970
The Collins conviction rested on two forensic pillars that the MSP Crime Laboratory’s analysts supplied at trial. Both have since been formally removed from valid forensic science.
Neutron Activation Analysis was presented as a method capable of identifying the elemental composition of hair samples with sufficient specificity to link a sample to a known source. MSP analysts testified that hair found in evidence was consistent with known samples connected to Collins and the Beineman crime scene. As Article 01 of this series documented, NAA has no accepted scientific basis for the attribution function it was assigned in the Collins trial. The National Academy of Sciences addressed this in its 2009 forensic science review.
Microscopic hair comparison was the second pillar. Analysts testified that hair samples were microscopically consistent. The DOJ’s 2015 review of hair comparison testimony in cases examined by the FBI found that analysts had overstated the significance of microscopic hair comparison in over 95 percent of trial transcripts reviewed. MSP’s testimony in the Collins trial followed the analytical conventions of 1970. Those conventions have since been established as scientifically invalid.
The lab did not commit fraud in 1970. It applied the methods that the forensic science community accepted and the courts admitted. The problem is not individual analyst misconduct. It is that an institution was trusted to supply determinative forensic testimony based on methods that did not do what they were presented as doing, and that conviction record has not been formally revisited in the Michigan Murders context.
The Contamination Record: Jane Mixer, 2002
The second failure is documented in the same laboratory that held the Collins evidence. In 2002, MSP’s forensic science division conducted DNA testing on physical evidence from the Jane Mixer case, in which Mixer had been murdered in 1969 and attributed to Collins as one of the uninvestigated six. The DNA analysis returned a contributor profile that was not Collins.
The profile was subsequently matched to John Ruelas, a man with no known connection to the Michigan Murders investigation or to Jane Mixer. Ruelas was convicted of Mixer’s murder in 2005 by Washtenaw County Circuit Court.
Physical evidence held in the MSP lab from 1969 to 2002 required an unbroken and documented chain of custody for any DNA result to carry evidentiary weight. The DNA analysis produced a result that led to a conviction. It also produced questions about what else the analysis found and how evidence held in a laboratory for more than thirty years maintained the integrity required for DNA analysis. The chain of custody record for Mixer case evidence in MSP custody has not been publicly documented in full.
The Mixer case is the clearest example in this series of a failure type that runs deeper than method error. In 1970, the lab used invalid science that was institutionally accepted. In 2002, the lab used valid science on evidence whose handling history over three decades introduced its own evidentiary instability. Both failures run through the same institution.
The Clutch Justice Lab hosts free investigative tools including the Michigan Murders forensic evidence timeline, the MSP lab failure tracker, FOIA generators, and court system reference documents. No signup required.
Access the Lab ?The Institutional Record: Backlog, Accreditation, and Digitization
The two failures documented above did not occur in an otherwise high-functioning institution. The MSP Crime Laboratory has an independent institutional performance record that provides context for both.
The lab’s documented backlog exceeded 19,000 cases at a reported point in its operational history. A backlog of that scale in a forensic laboratory is not an administrative footnote. It is a measure of the gap between the institution’s capacity and the demands placed on it, and it has direct consequences for case timelines, evidence integrity, and the accuracy of the analytical work the lab produces under pressure.
The lab also has a documented accreditation lapse in its record. Forensic laboratory accreditation is the external verification mechanism that provides baseline assurance that a lab’s methods, chain of custody procedures, and quality controls meet established standards. An accreditation lapse is documented evidence that those standards were not being met at the time of the lapse.
The MSP lab’s documented refusal to digitize older case records is the third institutional decision in this pattern. Physical evidence and case records from the 1960s and 1970s exist in analog formats that degrade with time, resist remote access, and cannot be systematically cross-referenced with newer evidentiary databases. A lab that holds the physical record of the Michigan Murders and declines to digitize it is making an active choice about what future reinvestigation can access. The record documents the refusal. It does not document the reasoning behind it.
The Lab as System
The Michigan State Police Crime Laboratory is the common institutional thread running through the entire Michigan Murders archive. It provided the forensic testimony that convicted Collins in 1970. It held the physical evidence for decades. It conducted the DNA testing that fractured the Mixer attribution in 2002. Its backlog, accreditation gaps, and digitization decisions have shaped what the investigative record contains and what investigators can access.
Each of these facts is documented in the public record. Taken individually, each can be characterized as an error, a resource constraint, or an administrative decision. Taken together, they describe a laboratory whose institutional architecture was systematically insufficient for the responsibilities it held.
This is the fourth article in the Michigan Murders Files series. Articles 01 through 04 have documented four distinct dimensions of institutional failure in the Michigan Murders archive: the junk science that framed the conviction, the six cases the prosecution chose not to try, the official network through which these decisions moved, and the laboratory that produced and held the underlying physical record. None of these is a standalone failure. They are the documented operating conditions of a system that closed two major connected investigations without closing them.
Forensic Method Failure Sorter
Tap each method to see what the lab supplied and why the later scientific record matters.
Neutron Activation Analysis
The Collins trial used NAA for attribution work it could not scientifically support. The later NAS record matters because it shows the method was treated as stronger than it was.
Microscopic hair comparison
Hair comparison was presented as meaningful identification evidence before the field had the empirical foundation to support that certainty.
DNA on old evidence
The Mixer evidence involved valid DNA technology applied to evidence that had been held for decades, making chain of custody and contamination central questions.
MSP Lab Risk Map
The article tracks more than one failure type. This map separates method error, evidence handling, and institutional capacity.
Reinvestigation Trigger Checklist
When these conditions appear together, old homicide evidence deserves renewed scrutiny.
The public record does not support a fraud finding. MSP analysts applied methods that were institutionally accepted and court-admitted in 1970. The problem is that those methods have since been established as scientifically invalid for the attributive function they were given. The failure is institutional and methodological, not individually criminal on the available record.
The 2002 DNA result identified John Ruelas as a contributor to physical evidence held in the MSP lab since 1969. The result led to a conviction. It also raised chain of custody questions about evidence stored in a forensic lab for more than thirty years. The chain of custody documentation for the Mixer evidence in MSP custody has not been fully disclosed in the public record.
A documented backlog of that scale indicates an institution operating beyond its processing capacity. For older cases held in that institution, it raises questions about priority, storage conditions, and whether evidence review for post-conviction purposes has been systematically deprioritized.
The public record documents that the forensic methods used in the Collins conviction have been repudiated. Collins has been incarcerated since 1970. Post-conviction review based on discredited forensic science is a documented legal pathway in Michigan and other states. No such review has been publicly filed or reported in the Collins case as of the publication of this article.
- National Academy of Sciences. Strengthening Forensic Science in the United States: A Path Forward. National Academies Press, 2009. Repudiation of NAA and hair comparison documented throughout.
- Department of Justice / FBI. Microscopic Hair Comparison Analysis Review. 2015. Analysis of hair comparison testimony in reviewed trial transcripts.
- People v. Collins, Washtenaw County Circuit Court (1970). MSP forensic testimony admitted by Judge John W. Conlin.
- People v. Ruelas, Washtenaw County Circuit Court (2005). Conviction based on 2002 MSP DNA analysis of Mixer case evidence.
- MSP Crime Laboratory. Accreditation and backlog documentation referenced in published reporting and state legislative oversight records.
- MSP Forensic Science Division. DNA analysis, Jane Mixer evidence (2002). Chain of custody and contributor identification documented in Ruelas trial proceedings.
Continue Your Investigation
If this reporting raised more questions, use the Clutch Justice ecosystem to keep going.
