Looking for information on the new rules taking effect on August 31? Click here!
WHAT INFORMATION DO YOU NEED TO SET ME UP FOR TESTING?
- the business name you are licensed under
- a principle contact person, phone number, and email (for results and invoices)
- your physical and billing address
- your UBI number
- and your WSLCB license number.
We never charge fees for account setup, and make no contracts requiring monthly/yearly subscriptions or minimum volumes.
HOW MUCH PRODUCT DO I HAVE TO SEND YOU FOR TESTING?
I-502 now requires a minimum of four (4) grams of flower material. (Be sure to read below “How Should I Select My Samples”).
Infused extract, inhalation extracts, kief, and hash require 1 gram.
Infused edibles, liquids, and topicals require 1 unit (serving/dose).
HOW SHOULD I SELECT MY SAMPLES?
There are many legitimate schools of thought on the issue of how to select unbiased samples from a “lot” of agricultural product. All of them share a common goal of selecting a sample that is representative of the lot in question. A representative sample is one that is composed of material that is similar in size, shape, density and overall quality to the majority of material in the lot from which it was selected. June 18th, 2016, the WSLCB will enact rule describing a sample selection protocol that they have determined to be adequate to implement as a minimum standard for our industry. The exact text of the WSLCB’s sampling protocols can be found on their website and in Washington Administrative Code 314-55-101.
Put simply, the WSLCB will be requiring that the person selecting the sample select not one, but four unique samples not less than one gram each taken randomly from all four quarters of the lot. Those four samples will be combined into a single sample not less than 4 grams, and the person who pulled the sample will initial their name in the traceability system stating that they selected randomly and representatively and did not alter or adulterate the sample.
A good sample selection protocol (one that produces consistent and representative results) is a difficult thing to achieve, and the sample selection step is the single greatest contributor to variance in test results. For the interested reader, click to read more.
The overarching theme of a good sample selection protocol is that it reasonably produces a sample which is most representative (most like the average) of the lot from which it is selected. No lot or batch of cannabis product is completely homogenous, and especially lots of green plant material can have a wide variance of potential values from a randomly selected sample.
It is safe to assume that the distribution of potential values from a population of randomly selected samples of the same lot is a normal distribution, meaning that most samples will produce values near the average and few samples will produce values near the extremes of the distribution. Still, it is possible – even likely when the distribution is wide – that a randomly selected sample of a lot will produce values that are not representative of the average of that lot (even worse if the sample is not selected randomly – that’s bias).
To overcome this problem, the sample selection protocol described by the WSLCB instructs the person selecting the sample to select one sample randomly from the lot not less than four (4) times and then group those four samples together as one. By selecting four distinct, yet similarly sized, samples randomly and then grouping the four samples together into one larger sample, we are able to minimize the amount of random error that might occur from selecting a single random sample. Furthermore, by increasing the minimum size of the sample (in grams) we increase the power of the analysis to accurately predict the average value of the lot.
Like any good sample selection paradigm for an agricultural industry, the one to be enforced by the WSLCB begins by describing appropriate hygienic practices. Because microbiological and foreign matter screening are endpoints to the testing regimen, it is important to maintain sanitary conditions of all equipment and containers used to remove, weigh, and transport samples. It is important that the persons handling samples wash their hands before doing so and remove from their hair, clothes, and bodies excess dust and debris that may contaminate the sample. Use of sanitary and powder-free gloves such as latex, nitrile, or vinyl is a requirement. Any tools used to remove the sample from the lot must be sanitized. Any container used to transport the sample (baggies, jars, etc.) should be in new condition or have been sanitized prior to sample addition. Any scales or other measuring devices that are to come in contact with the sample must also be sanitary.
Sanitization of tools and instruments is most easily done with a wipe from a clean paper towel soaked with 70% Isopropyl Alcohol and then wiped again with a clean and dry paper towel until the utensil is dry. All visible dust and debris should be removed. Sterilization by flaming is not recommended for this application.
To select a sample from a flower lot, the lot in question must be quartered; the lot is divided into four equal quarters each representing ¼ of the lot. Quartering of the lot can be done physically or visually (i.e. eyeballing it). Physical quartering of the lot is most easily done by laying the lot out on a sanitary surface in a pile not more than two inches tall in any place. Using a sanitary ruler or yardstick separate the pile down the middle then separate both halves again down their middles to produce four non-touching quadrants. The quadrants do not need to be of the exact same weight, but they should be of roughly the same weight. Visual quartering is most easily done by standing over the container housing the lot and painting a mental image of an X centered on the container and extending beyond the container’s borders. Once the lot has been quartered, one sample weighing not less than 1 gram is to be selected randomly from each of the four quadrants.
If visually quartering, samples should not be taken always from the top of the container, but all depths of the container should be equally likely to be selected from. Each of the four samples taken are then combined into the same sample transport container to create a single sample not less than 4 grams in total weight of contents.
A truly random sample selection strategy is one in which all potential samples in the lot have an equal chance of being selected. While this requirement may seem simple at first, achieving a truly random sample selection is challenging. For instance, if visual quartering is employed on a lot housed inside a large bag or tub, and samples are selected only from the top layer of the container, then samples at the bottom do not have an equal chance of being selected. Furthermore, if some buds in the lot are greater than 7 grams they cannot be selected as samples unless they are broken because the maximum sample size in i502 currently is 7 grams.
In other agricultural industries regulated by the FDA, USDA, or EPA, a variety of randomization techniques are employed/required. These protocols are often complicated and highly involved, and the exact requirements vary depending on type of product and its common storage and processing states. Because of the industry’s infancy, such rigorous protocols for random sample selection from cannabis flower lots have not been defined. At this point in time, the regulatory authorities are relying on individual processors to define their own randomization strategies. We at Confidence Analytics would be happy to work with processors to create and describe sampling protocols appropriate for their workflow to further improve this important step in the process. If your operation is interested in furthering the cutting edge in this regard, please give us a call.
HOW SHOULD I PACKAGE MY SAMPLES FOR TESTING?
Package each sample separately. Plastic baggies work great for flower, kief, trim, and edibles. A plastic container with a lid is best for waxy or sticky products, such as hash and some extractions. Syringes are also very good containers for extractions that are not too thick for them. In order to expedite processing we ask that you include two barcodes on your package, the QA sample number from BioTrack and the number of the lot the sample represents. If possible try to use a barcode printer that does not put a black line box around the barcode; while BioTrack has listed some of those printers as compatible, there are no barcode readers commercially available that can actually scan these barcodes. An inexpensive Dymo printer is a good option for barcode printing as an alternative. If, for some reason, you are not able to barcode your products at the time of drop off, please include these two numbers somewhere on your packages. Ideally, we’d also like to know the lot’s harvest date and it’s sativa/indica percentage to the best of your knowledge.
DO I HAVE TO PUT MY PRODUCT INTO QUARANTINE?
When you make your QA samples, the traceability software may say they have to be placed in quarantine for 24 hours. The WSLCB has told us directly that QA samples do not have to be quarantined.