The State Says Affinity Skipped a Safety Test on 78 Products. It Approved Every One.

The State of Connecticut has summoned the cannabis manufacturer MCEJV, LLC, which does business as Affinity Grow, to answer a charge that it sold 78 gummy products without the residual-solvent testing state rules require.

The Department of Consumer Protection’s Notice and Particulars, dated October 22, 2026, names every product.¹ It covers registrations from September 19, 2024 through December 18, 2025. On the 76 products in the main list, the Department writes:

“When registering these cannabis products, Respondent disclosed that each one was created using a solvent. None of them, however, were tested for residual solvents as required by Department Policy and Procedure §21a-421j-29(a), as indicated by the respective Certificates of Analysis. These above-listed cannabis products were subsequently sold, transferred, transported, or otherwise conveyed to retailers, hybrid retailers, and a medical-only dispensary facility.”

Two further products, registered on November 14 and 15, 2024, are handled separately. For those, the Department says the certificates claimed the testing had been done:

“The respective Certificates of Analysis, however, while stating that residual solvent testing was completed, showed no actual evidence or results of such residual solvent testing.”

And then the line that carries the case:

“The cannabis testing laboratory that performed the Certificate of Analysis testing on these cannabis products confirmed that Respondent specifically clarified that solvent testing was not needed, that no such residual solvent testing was performed, and also noted that it appeared Respondent had registered the products with the incorrect Certificates of Analysis.”

The Department says it has received no evidence that the testing was ever performed on those two.

A hearing is set for December 8, 2026 at 10 a.m., by Microsoft Teams, to show cause why the company’s licence should not be revoked or suspended. The allegations are disputed and no decision has been made.

What the company told consumers

Connecticut requires cannabis packages to disclose all processing techniques and solvents used, and permits producers to deliver that through a QR code.²

Affinity Grow’s QR code leads, through a commercial shortlink, to a folder the company maintains on Dropbox, organised by product category, then year, then individual product. Inside the folder for product 00204 — Chew Pineapple Toasted Coconut Gummies S, registration MMBR.0024885, one of the two products the Department singled out — is a one-page PDF dated December 3, 2024.

It gives the manufacturing date, the nutrition panel and the allergen statement. It also says this:

“The following solvents have been used in the production of this product: Ethanol / Heptane”

Directly above it, the same document says what was checked:

“This product has been tested for microbiological contaminants, mycotoxins, heavy metals and pesticide chemical residue.”

Four contaminant classes. Residual solvents are not among them.

So the company’s own consumer disclosure names two solvents it used, and the testing statement on the same page does not include solvent testing. That document was published by Affinity, not by the state, and it is not referenced in the Department’s notice.

What the public filings show

Connecticut CannaTimes retrieved all 78 certificates named in the notice from the state’s public brand registry and checked each one.³

All 78 registry records declare that a solvent was used. Seventy-five of the certificates contain no residual-solvent panel at all. Two contain a panel headed “Complete” above an analyte table with nothing in it — no analyst, no instrument, no date tested. One contains a real panel with a result, on a certificate regenerated on January 29, 2026, six weeks after the product it supported was approved.

Sixty-six certificates came from Northeast Laboratories, twelve from Analytics Labs CT.

Every figure in this section was recounted on October 25, 2026 by retrieving all 78 certificates again from the state registry. The retrieval manifest lists each one with its registration number, the laboratory, what its residual-solvent panel shows, the SHA-256 of the file retrieved, the state URL it came from and the date it was retrieved.

The Department approved registration MMBR.0024885 on November 14, 2024 — the day after its certificate was generated, and the day before it approved the second one.

Two Affinity certificates, one of them among the charged 78 and one not, were replaced on the state registry after their products were already approved. Each says, printed on the replacement, “Updated to add residual solvents.” One was replaced eight days after approval, the other forty-two days after. Both replacements postdate the first compliance meeting in this case. They are the only replacements of that kind this publication identified among the certificates it reviewed. The superseded versions are not recoverable from the state’s website, and nothing in Connecticut’s rules requires them to be kept.

The Department’s evidence is the document it already approved

Read the notice closely and a second question appears, one the Department does not address.

The charge against the 76 products is proved, in the Department’s own words, “as indicated by the respective Certificates of Analysis.” For the two November 2024 products, the Department says the certificates state that residual-solvent testing was completed while showing “no actual evidence or results of such residual solvent testing.”

Those certificates were not obtained by subpoena. They were not recovered from the company’s files. They are the documents Affinity filed with the state when it applied to register each product, because the rules require a certificate with every application. The Department received all 78. It approved all 78.

The defect the Department now describes is on the face of page four of the two November 2024 certificates. Under the heading “Residual Solvents” the status reads Complete. Beneath it: “Entered by:” blank. “Instrument ID: 0.” “Date Tested:” blank. Then the column headers for analyte, limit, result and status, and nothing under them.

The state’s public registry shows MMBR.0024885 was approved on November 14, 2024, the day after the certificate’s PDF was generated. MMBR.0024892 was approved the following day.

So the Department approved both registrations, and then, at some later point, read the same two documents and found what was wrong with them. Either nobody looked the first time, or somebody looked and approved them anyway. The record does not say which, and the Department has not explained it.

There is a rule that makes the first possibility unremarkable. The grounds on which the Department must refuse a product registration run to eight, and every one of them is about the product’s name — duplicate, confusingly similar, obscene, appealing to minors, unsupported health claims. Testing is not among them. The Department’s own registry page says a listing “only indicates approval of the product’s brand name” and “is not a representation of legal compliance.”

The Department can say, accurately, that nothing obliged it to open the attachment.

It approved three more after it knew

Whatever happened in November 2024, the Department was on notice by the summer of 2025. Its own revised notice says that on July 1, 2025 it reviewed Affinity’s surveillance footage in the course of this investigation.

Connecticut CannaTimes reviewed the gummy certificates filed with the state registry during the charged window. Four certificates we can produce carry the signature the Department charged against the two November products: a residual-solvent panel headed “Complete” above a results table with no values in it. All four were issued by the same laboratory, Analytics Labs CT. Each certificate discussed in this article is published in full in the records archive, and the retrieval manifest lists every one with its registration number, the state URL it came from, its SHA-256 and the date it was retrieved.

The two uncharged certificates are not identical to the charged pair, and the difference matters. On MMBR.0024885 and MMBR.0024892 the panel is wholly unpopulated: “Entered by:” blank, “Instrument ID: 0,” “Date Tested:” blank. On MMBR.0028562 and MMBR.0028831 those fields are filled in — an analyst code, an instrument identifier and a date tested — and only the result rows are absent. The two records differ in what they show about whether anything was run.

The Department approved the two uncharged registrations on July 30 and August 19, 2025 — within seven weeks of the date it says it was reviewing this company’s footage. Both are registered for both the medical and adult-use markets.

Those two companies are accused of nothing, by anyone. A registry declaration that a solvent was used is a self-disclosure, not a legal determination that a finished-product panel was required for that product. What the dates describe is the Department, not them.

Across every certificate reviewed for this investigation, not one carries a digital signature or any other integrity protection. When a certificate is amended, the replacement silently overwrites the old one on the registry and no version history is kept. Connecticut’s rules contain no provision at all governing the amendment or reissue of a certificate of analysis.

The first incident

The notice contains a second set of allegations, and they are different in kind.

The Department says that on June 4, 2025, Affinity recorded in the state’s seed-to-sale tracking system that cannabis flower of the strain CAP JUNKY had been converted into 1,294 units of a product called 5DCJ62-PR-3. A week later, it says, the company created one additional unit, then split that unit through a series of adjustments into five sublots of one, one, one, three and three units, all still identified as CAP JUNKY. Minutes later, the Department says, the company renamed each sublot.

On July 1, 2025, according to the notice, the Department reviewed the company’s surveillance camera footage and confirmed that the five sublots were, in reality, five separate and unrelated products from five separate strains: Bachelor Party, Zangria, Soap, Gorilla Glue #4 and Medellin.

The Department also alleges the company failed to correct the entries and notify it within twenty-four hours.

For Incident 1 the Department charges false, misleading or deceptive representations;^1,5 involvement in fraudulent or deceitful practices; failure to keep accurate records; failure to comply with RERACA and the Department’s policies and procedures; and two tracking-system provisions. For Incident 2 it charges failure to have all cannabis receive a satisfactory certificate of analysis, fraudulent or deceitful practices, and failure to comply.

Connecticut CannaTimes searched the state’s brand registry for CAP JUNKY.³ It appears nowhere — not under Affinity, not under any producer, on any date. The five real strains all have many registrations before and after June 2025.

This incident is what separates Affinity from every other Connecticut producer. No comparable allegation against another Connecticut manufacturer appears anywhere in the public record — and Connecticut publishes no register of cannabis enforcement actions, so the public record is the limit of what can be checked.

The company’s history with the Department

Affinity’s record with the Department runs in both directions, and the state has been found at fault in part of it.

In October 2024 the Department ordered a recall of seventeen lots over a sampling dispute: whether a laboratory employee had to physically remove a sample from a batch or merely select it. Some lots were too small to reprocess and were destroyed. The alleged procedural failure was the laboratory’s. The penalty fell on the cultivator, and no public action against the laboratory appears in the record.

In March 2025 the company’s chief executive testified against the Department’s position before the General Law Committee.⁶ Roughly twenty-four hours later the Department’s Drug Control Division director ordered an unannounced compliance check. The inspection found the company was testing compliantly and the case closed. The commissioner apologised publicly and said the check took place without his knowledge or approval. The state’s own labor relations investigation found the director’s judgment detrimental to the agency; he resigned that August.⁷ In April 2025 the Department removed cannabis oversight from the Drug Control Division entirely.

The state’s investigation file records eighteen complaint-generated inspections and five regular inspections of this one company.

The company is seeking permission to sue the state for $1 million over losses it attributes to the Department’s handling of its regulations. The claims commissioner rejected the state’s motion to dismiss.

The company has said publicly that the current dispute is about the science, that the state’s chromium testing method is not reliably enforceable, and that it commissioned an independent study saying so.

What is not established

Nothing in this record shows that any Affinity product contained residual solvents at any level, or was unsafe. Missing testing is missing information, not a detection.

The record does not establish who set the “Complete” status on the two empty panels, when the laboratory learned of the problem, or whether it told anyone. Connecticut’s rules impose no duty on a testing laboratory to report that a client declined a required test.

It does not establish why the two certificates were amended. The innocent reading is that the laboratory ran the test, corrected the record, and the Department’s charge list reflects the correction.

Registrations are not batches, units or packages. Seventy-eight registrations is seventy-eight registrations.

Every allegation in the notice is disputed and unadjudicated.

So who is checking

The structure the record describes has three parties in it.

The manufacturer has a clear duty. Department Policy and Procedure §21a-421j-29(a) bars conveying cannabis to a retailer without a certificate confirming satisfactory results for all required testing, and solvent testing is one of those tests for anything made with a solvent.⁸ That is what the Department charged.

The laboratory sits in the middle. The notice says it confirmed to the Department that Affinity clarified solvent testing “was not needed, that no such residual solvent testing was performed.” Connecticut’s reportable-events rule binds a licensee “and any employee, including a cannabis testing laboratory employee” to notify the Department immediately, in writing, of any suspected diversion, theft, loss, adulteration or unauthorized destruction of cannabis, or loss or unauthorized alteration of records.⁸ A client declining a required test is none of those. No Connecticut rule requires a laboratory to report it. No public enforcement action against any Connecticut cannabis testing laboratory could be located; the state publishes no enforcement register.

The Department received a certificate with every registration, and approved every registration. Nothing in the rules required it to read any of them.

Three parties. One safety test the Department says was skipped on 78 registrations by one company. Another 222 registrations statewide whose filed certificates carry no residual-solvent results at all, which is a finding about documents and not a finding that those products went untested. And rules that place a clear obligation on exactly one of the three.

No Connecticut statute requires that a finished cannabis product be tested for residual solvents. Public Act 26-8, the 232-page cannabis act the legislature passed this year, refers to a solvent or solvents eight times, and every one of those references governs which solvents a processor may use — a Class 3 solvent within the meaning of United States Pharmacopeia Chapter 467, a solvent generally recognized as safe, or a solvent the commissioner approves and posts.⁹ Not one of them requires a test. The statutory testing provision in force during the charged window named microbiological contaminants, mycotoxins, heavy metals and pesticide chemical residue, and did not name residual solvents.⁴

The residual-solvent requirement exists only in the Department’s Policies and Procedures, a document that is not a regulation and expires on July 1, 2028.

Every allegation in the notice is disputed and unadjudicated, and the hearing is December 8.

What the record does show is that on 77 of the 78 certificates, the residual-solvent results the rules require are not on the document. Those certificates were filed with the state 78 times, and the state approved all 78 before it acted on what was in them. Whether the testing was performed is the question the December 8 hearing exists to answer.

Corrections

October 25, 2026. Four changes, all to the statewide section.

The article reported a census of “597 products, all ten licensed producers” and said five certificates statewide carried the charged signature. That census was carried out, but the file recording it was not preserved and this publication cannot now produce it on request. A figure we cannot show a reader is a figure we should not print. Both have been removed, and the section now states only what the certificates we can produce show. The census also matched products by the word “gummy” in the registered name, so solvent-declared edibles named otherwise fell outside it — at least one charged product did.

A fifth row, MMBR.0028240, registered to Connecticut Contract Manufacturing LLC, has been removed. The reading was made against the certificate as the state served it, but no copy was kept and the registry link now returns an error. Nobody can check it, so it should not stand. No adverse inference should be drawn about that company.

The table named Alpha CT 4 LLC as the producer of MMBR.0028562. No entity under that name appears in the state’s product registry, which records the branding entity as SHANGRI-LA CT INC. The table now names the entity as the registry does.

The uncharged certificates were described as carrying “the same signature” as the two the Department charged. They do not. The charged two have no analyst, no instrument and no date tested; the uncharged ones have all three and lack only the result rows. A passage has been added setting out the difference.

October 24, 2026. An earlier version of this article stated in its closing paragraph that a required safety test “went unperformed on 78 products” and that the certificates proved it. That overstated the record in two ways. Whether the testing was performed is a disputed allegation the Department will have to prove at hearing, not an established fact; and one of the 78 certificates does carry a residual-solvent result, added by amendment in January 2026, so the absence of results appears on 77, not 78. The paragraph has been rewritten.

The same version described the 222 statewide registrations as instances of the same missing safety test. The 222 figure counts certificates filed with no residual-solvent panel. That is a finding about documents, and it is not a finding that those products went untested. The passage has been rewritten to say so.

A paragraph in the first-incident section discussed chromium and the heavy-metals testing requirement under Public Act 24-76. The Department’s notice does not allege a heavy-metals or chromium failure and does not mention either. The paragraph has been removed.

Three statements that no enforcement action had been taken against another manufacturer or against any testing laboratory have been changed to say that none could be located in the public record. Connecticut publishes no register of cannabis enforcement actions, so absence from the public record is not proof of absence.

An earlier version stated that the phrase “residual solvent” does not appear in Public Act 26-8. It appears once, in a provision governing which solvents may be used to extract hemp. The act refers to a solvent or solvents eight times in total and none of those references imposes a testing requirement, which is the point the sentence was making; the sentence has been rewritten to state it accurately.

The document

The Department’s revised Notice and Particulars of October 22, 2026, five pages, is published in full in the Connecticut CannaTimes public records archive: Revised Notice and Particulars, DCP Case 2025-978. It is posted exactly as the Department issued it, with no redactions by this publication, and it is full-text searchable.

Sources

1. Revised Notice and Particulars, In the matter of MCEJV, LLC, License No. ACME.0008473, Case No. 2025-978, Connecticut Department of Consumer Protection, October 22, 2026. Published in full: https://connecticutcannatimes.com/documents/ctc-2026-0408/
2. Labeling Requirement for Processing Techniques and Solvents, Connecticut Department of Consumer Protection licensee notice, January 5, 2024. https://portal.ct.gov/cannabis/-/media/cannabis/notices/2024/1524-labeling-requirement-for-processing-techniques-and-solvents.pdf
3. Medical Marijuana and Adult-Use Cannabis Brand Registration, Connecticut Open Data, dataset egd5-wb6r. https://data.ct.gov/Health-and-Human-Services/Medical-Marijuana-and-Adult-Use-Cannabis-Brand-Reg/egd5-wb6r
4. Public Act 24-76, Substitute House Bill No. 5150. https://www.cga.ct.gov/2024/act/pa/pdf/2024PA-00076-R00HB-05150-PA.pdf
5. Connecticut General Statutes section 21a-421p. https://law.justia.com/codes/connecticut/title-21a/chapter-420h/section-21a-421p/
6. CT Mirror, cannabis inspection and the Department of Consumer Protection, April 9, 2025. https://ctmirror.org/2025/04/09/ct-cannabis-inspection-consumer-protection/
7. State of Connecticut investigation summary, Drug Control Division director, May 30, 2025. https://npr.brightspotcdn.com/5d/e3/07d366964902a14683634c982ae4/investigation-summary-marriott-rodrick-5-30-25.pdf
8. Adult Use Cannabis Policies and Procedures, effective November 12, 2024, sections 21a-421j-2 and 21a-421j-29. Connecticut eRegulations System, operational policies. https://eregulations.ct.gov/eRegsPortal/Browse/getOperationalPolicies
9. Public Act 26-8, Substitute House Bill No. 5350. https://www.cga.ct.gov/2026/act/Pa/pdf/2026PA-00008-R00HB-05350-PA.PDF