Tampa International Airport police seized fifty vacuum-sealed bundles of a green, leafy substance from two duffel bags in July 2022. The Florida Department of Law Enforcement (FDLE) lab tested one of them. It was cannabis, and a Hillsborough County jury convicted the owner of trafficking more than twenty-five pounds. In March 2025 the Second District Court of Appeal threw out the trafficking conviction in Campbell v. State, No. 2D2023-0651 (Fla. 2d DCA Mar. 26, 2025), and sent it back as a possession case.
One tested bundle couldn’t prove what was in the other forty-nine. Hemp has been legal in Florida since 2019, and what separates it from illegal cannabis is how much THC, the plant’s main intoxicating compound, it holds. The analyst testified that her first two tests, a microscope exam and a Duquenois-Levine test (a reagent that changes color around cannabis), can’t tell hemp from marijuana because neither one measures THC. Officers couldn’t tell by look or smell either. The weight that made it trafficking rested on forty-nine bundles nobody tested. The State wanted the jury to read one bundle as fifty.
That gap, between what a lab actually tested and what the report gets used to prove, is where a lot of Florida drug cases turn. A report naming a drug and a weight is a summary. The records under it show which pieces were tested, how and on what equipment.
Which Bag Did the Lab Actually Test?
Labs rarely test everything, and Florida law allows that: section 893.105 says seized drugs “may be sample tested and weighed by the seizing agency,” and the results come in to prove “the nature, composition, and weight” of what was seized. The fight is over how far one test reaches.
The national standard for seized-drug labs, the SWGDRUG Recommendations from the Scientific Working Group for the Analysis of Seized Drugs (Version 8.3, September 2026), draws the line clearly. A lab can pick what to test by a statistical plan, which lets it state a conclusion about the whole seizure at a stated confidence, or by a non-statistical choice, which generally can’t speak for the units nobody tested. When the choice isn’t statistical, the report has to say the result covers only the tested units, and SWGDRUG supplies the sentence: “2 of 100 bags were analyzed and found to contain Cocaine.”
Florida’s courts got there first for powders. In Greenwade v. State, 124 So. 3d 215 (Fla. 2013), nine baggies of white powder were emptied together and tested as one. The Florida Supreme Court held that before individually wrapped packets of white powder can be added together for a trafficking weight, the State “must chemically prove that each individually wrapped packet contains at least a mixture of a controlled substance.” The Court limited the rule to substances with a real danger of being misidentified, which at the time left out marijuana and rock cocaine. Campbell held that legal hemp put cannabis back in.
What a Test Has to Include to Count
Even the one bundle that gets tested has to be tested the right way. SWGDRUG sorts lab techniques into three tiers by how well they tell one chemical from another:
| Tier | Examples | What it gives the analyst |
|---|---|---|
| Category A | mass spectrometry, infrared spectroscopy, Raman spectroscopy, NMR, X-ray diffraction | structural information about the molecule |
| Category B | gas chromatography, liquid chromatography, ion mobility spectrometry, microcrystalline tests, thin layer chromatography | chemical and physical characteristics, without structure |
| Category C | color tests, immunoassay, melting point, pill markings | general or class information |
One test is never enough. With a Category A technique, the lab needs at least one more that relies on different properties of the drug. Without Category A, it needs at least three, two of them from Category B. A color test, in the field or at the bench, is Category C. So are pill markings, because counterfeits copy them.
GC-MS, the instrument most drug labs lean on, is two techniques joined together: the gas chromatograph separates the mixture, and the mass spectrometer reads each piece. SWGDRUG lets a lab count it as two only when both results meet the lab’s acceptance criteria, and it requires data another chemist can review, such as the spectra and chromatograms. The mass spectrum usually gets matched against a reference library, and the software ranks the closest entries by score. A high score points toward an answer. What turns it into an identification is a result that meets the lab’s written criteria, checked against a reference material or reference data. The report won’t show any of that. The data will.
Weight Sets the Sentence
Florida’s trafficking statute counts the weight of “any mixture” containing the drug, not its purity. For cocaine, 28 grams of a mixture starts trafficking with a three-year mandatory minimum (section 893.135(1)(b)1.a.). Near a line like that, the lab’s balance matters as much as its mass spectrometer.
Every weighing carries some uncertainty, and an accredited lab has to work it out. In Campbell the tested sample was reported as 24.47 grams “plus or minus 0.15 grams.” When a weight sits close to a threshold, I want the balance’s calibration checks around the test date, whether the weight included packaging, and how the stated uncertainty was calculated.
The Records Under the Report
In a Florida case these come through discovery under Rule 3.220, a subpoena or a public records request, and the charge decides which ones are worth the fight:
| Record | What it can show |
|---|---|
| The sampling plan and the analyst’s notes on what was opened | which units were tested, and whether any were combined first |
| The analytical scheme for the exhibit | which techniques were used, and whether they meet the lab’s own minimum |
| Spectra, chromatograms and the library hit list with scores | whether the data supports the identification |
| For plant material, any THC measurement | whether anything separates the sample from legal hemp |
| Balance calibration and the weight’s uncertainty | how close the reported weight is to a trafficking line |
| The electronic audit trail | re-runs, re-integrations and edits to the data, with dates and names |
| The analyst’s proficiency tests and the lab’s accreditation scope | whether the method and the person were checked around the test date |
One more question belongs on the list: does the analyst on the report still work at the lab? In Smith v. Arizona, 602 U.S. 779 (2024), the analyst who tested seized drugs had left, and a substitute told the jury her results. The Supreme Court held that an absent analyst’s statements relayed to support an expert’s opinion are offered for their truth, so the right to confront that witness can’t be sidestepped by calling the testimony an independent opinion.
The accreditation records have their own page, on lab accreditation. The difference between a presumptive result and a confirmed one is on the screening versus confirmation page, and questions about software in the lab are in the post on AI in the crime lab. If you’re charged with trafficking or another drug offense in Tampa Bay, the first thing I want to know is what the lab actually tested, and the challenging the drug evidence page covers the rest of that fight.
Two Questions About Lab Results
Can a Florida crime lab drug test be wrong?
It can, and the risk is highest where a lab skips steps the science calls for. The national seized-drug standard requires a combination of techniques before a lab names a drug, and a color test alone only gives class information. In 2025 a Florida appeals court threw out a trafficking conviction because the lab tested one of fifty bundles and its first tests couldn’t tell hemp from marijuana.
Does the lab have to test every bag in a Florida trafficking case?
For individually wrapped packets of white powder, each packet has to be chemically tested before the weights are added together. The Florida Supreme Court held that in Greenwade v. State (2013), and in 2025 the Second District applied the same reasoning to cannabis because legal hemp looks and smells the same.
The Opinions, the Statutes and the Standard
- Campbell v. State, No. 2D2023-0651 (Fla. 2d DCA Mar. 26, 2025).
- Greenwade v. State, 124 So. 3d 215 (Fla. 2013).
- Smith v. Arizona, 602 U.S. 779 (2024).
- SWGDRUG Recommendations, Version 8.3 (September 1, 2026), Parts III A and III B.
- Section 893.135, Florida Statutes, and section 893.105.
- Section 581.217, Florida Statutes, the state hemp program.
This post is general information about the law, not legal advice, and it does not create an attorney-client relationship. Past results do not guarantee a similar outcome. The law discussed is sections 581.217, 893.105 and 893.135, Florida Statutes; Campbell v. State, No. 2D2023-0651 (Fla. 2d DCA Mar. 26, 2025); Greenwade v. State, 124 So. 3d 215 (Fla. 2013); and Smith v. Arizona, 602 U.S. 779 (2024). The laboratory standard discussed is the SWGDRUG Recommendations, Version 8.3 (2026).

