When Uncle Ike’s Pot Shop opened in Seattle, owner Ian Eisenberg said he couldn’t compete with medical dispensaries’ lower prices, but he did have one advantage.
“Ours is tested and you know what you’re getting,” Eisenberg said.
He’s right on the first point. State rules require a small sample tested from every lot of marijuana up to five pounds.
But do consumers know what they’re getting? That’s murkier. As the state market develops, so does its testing program.
The program is having success screening substances like yeast, mold and bacteria. About 10 percent of marijuana buds fail tests and can’t be sold in recreational pot stores, according to Liquor Control Board data.
Potency testing, meanwhile, shows Washington’s recreational pot is all over the map. It averages about 16 percent THC, but ranges widely. About 2.5 percent of marijuana tests above 28 percent THC. Some samples climb into the 30s and 40s.
For perspective, High Times reports the “heaviest-hitting strains” at conventions it hosted in 2013 maxed out at 28 percent. In Colorado, scientists at CannLabs said they require a retest for results higher than 27 percent THC.
Does that mean Washington is growing some of the world’s strongest weed?
Most Read Stories
- Submarines dismantled in Puget Sound are symbols of nation’s defense dilemma | Jon Talton
- Democrats are supposed to be fighting back, but they just keep losing | Danny Westneat
- Spike Lee posts, then deletes photo thanking Seahawks' Pete Carroll for signing Colin Kaepernick
- Swedish double-booked its surgeries, and the patients didn't know | Quantity of Care
- Seattle Zestimates are off by $40,000; now hundreds of data crunchers vie to improve Zillow’s model
It could. But some laboratory directors, pot growers and the state Liquor Control Board have reservations about the early data and seek changes to the testing system.
In fact, laboratory leaders said they are forming working groups to lobby the LCB for more oversight of lab methods.
“Part of it is to invite more regulation,” said Brad Douglass, the scientific director at the Werc Shop, one of the 12 labs approved by the state.
Randy Simmons, the state’s marijuana project director, said the system is off to a good start but needs tweaking.
“The majority of what’s out there on the packages is correct,” he said.
He expects changes early this year.
“The lab side is emerging,” Simmons said. “As it matures, I think all those things that have been missed … or things we find out we should be looking for, will all be changed.”
The Werc Shop lab in Bellevue is a simple, white-walled office with a collection of expensive machines that look like fancy printers.
Marijuana samples sealed in sterile plastic bags are separated for several tests, including the state-required microbial and potency tests. Together, the tests cost $150.
For microbial testing, samples are fed into an automatic incubator, which takes up to 56 hours to grow and automatically test for salmonella, mold and other harmful substances.
For a potency test, pot is ground, weighed and placed into a tube with solvent that allows chemical compounds to separate as the tube spins in a centrifuge. Later, plant material is filtered out and the remaining liquid gets injected into a machine that analyzes the cannabinoid compounds.
THC, notably, gets you high. CBD is credited with many of pot’s medicinal benefits.
The Werc Shop’s process relies on automated equipment for testing. It’s built to minimize human error.
Much of that precision, though, is unraveled before samples even arrive, because what growers send might not reflect the overall crop.
“If you’re not careful with sampling, you can have results that vary greatly,” said lab manager Cameron Miller.
But LCB rules don’t direct growers on how to choose samples.
A grower harvesting a crop of 25 plants can select for testing the best-looking bud encased in THC-loaded crystals, rather than a brown one that looks like a hairball.
That’s something the LCB’s Simmons hopes to change.
Retailers often use potency results as a way to advertise the quality of their marijuana. Higher test results can mean higher wholesale prices for growers.
Some producers think others might be falsely boosting test results by adding THC-laden substances to their samples.
“I have suspicions some people are … rolling it in kief and getting high scores,” said Joby Sewell, whose company, AuricAG, grows marijuana in Sodo. (Kief is the powder made from glands that have been sifted or rubbed from the buds and leaves of the marijuana plant.)
Simmons said the LCB also is concerned growers could dip their buds in hash oil before tests.
The agency is beginning a secret-shopping program to find out. Simmons said incognito marijuana-enforcement agents will buy products from stores and have them tested. If the results don’t hew closely to the label, the LCB will investigate.
“Will we see people play games? Yes. It happens in any industry out there. Will we catch them? Yes, we will,” said Simmons.
Simmons said marijuana officers are investigating several producers. So far, no violations have been recorded.
Labs also are under scrutiny.
“There are some labs that have financial incentive to game the system for clients,” said the Werc Shop’s Miller.
A fundamental conflict resides at the heart of the testing industry. Producers pay labs for their tests. If producers don’t like the results, they could take their business elsewhere.
“In the medical world … people that do tests sometimes do pay for higher test results,” said Simmons. “I want to make sure that’s not happening on the recreational side.”
The Werc Shop’s Douglass said secret shopping seems premature because labs don’t have to use the same methods to test, though they are expected to conform to herb-industry standards adopted by the LCB.
Douglass wants the LCB to send the same sample to several labs, to check for consistency, and help identify labs with equipment or process problems.
Dr. Michelle Sexton, the chief science officer at state-approved PhytaLab in Kirkland, who helped edit the testing standards adopted by the LCB, said five different lab tests right now likely would yield “five different answers until we go to the next level and get validated methodology.”
That doesn’t mean labs are doing a bad job or fudging results. The expensive equipment The Werc Shop uses is not required. For a microbial test, Douglass said, labs looking for a “shoestring budget” alternative to his automated incubator might use a petri dish to grow bacteria and then manually count colonies with a microscope.
“Whenever humans are involved, there are going to be errors,” said Douglass.
In addition to secret shopping, Simmons said, the LCB is flagging outlying test results and will perform compliance checks on labs.
“You can do a different process,” said Simmons. “If it comes out differently from ours, you can’t use it.”
Independent tests question the numbers reported to the LCB.
Jessica Tonani, the CEO of Verda Bio, a Seattle cannabis biotech company, bought pot from recreational stores and had The Werc Shop blind-test them. Seven of eight purchases tested between 3 percent and 7.5 percent different from what their labels showed, according to Tonani’s data.
Small differences can be expected because plant material varies. How much does the LCB allow? Simmons said the agency hasn’t decided yet, but said 1 percent sounded about right.
“I think there are very valid scientists doing testing out there,” said Tonani. “But if your lab is high or low, there’s no data to show your lab is testing correctly.”
“Stronger all the time”
Although it’s an imperfect system, don’t discount the idea that Washington could be growing increasingly powerful weed.
“It’s getting stronger all the time,” said the LCB’s Simmons. “People are growing hybrids to increase the THC content.”
Dr. Jonathan Page, an adjunct professor at the University of British Columbia and founder of a Canadian cannabis biotech company, said it’s possible that “better growing conditions, as would be expected in a legal place, and good genetics are coupling together to push it beyond what we thought is the limit.”
Page pointed to a study of pot seized by police in Australia. More than 200 samples were tested. They averaged about 15 percent THC, similar to Washington. About 7 percent of the Australian pot tested above 28 percent. One sample hit 39.8 percent.
Is high-THC pot better? Not to Page, who sees it as “the Everclear alcohol of the marijuana world.”
Page said early research shows that more than 100 minor cannabinoids (not THC or CBD) play a role in pot’s effects.
“What I’d like to see is that there’s less of a focus on THC as sole molecule, and people saying, ‘I have the most potent bud,’ and more focus on cannabinoid profile” and the overall user experience, said Page.
That might mean more focus on terpenes, which are compounds that create flavor and smell in pot.
Said Douglass: “As consumers become more educated, they’ll be able to sense terpenes in cannabis like a oenophile does with wine.”