6 Likes
Oh also I saw recently that you can batch test seeds for gender which is kind of neat.
Elaborate?
If you’re dealing with feminized seed and you just need a ratio?!?
1 Like
It was part of some joint marketing research thing. I’d have to dig it up for the specifics. High level the idea is you can test a handful of seeds in a single batch to determine if its all one gender or a mix. So yeah, if you’ve got 100,000 hemp seeds you want to make sure are feminized, they’re pitching you can test before planting.
Edit: Seed Gender Testing Study.pdf (2.1 MB)
2 Likes
Total statistical analysis via ANSI AQL sampling. It works for most things that potentially have a normal distribution on a bell curve. 
2 Likes
It’s not that it’s harder, more about the labor and cost involved in going the natural rout of space, coco, nutes, 6weeksish waiting compared to 10 days from sprout. I keep every male anyway (for breeding purposes) just not the same as the females obviously, less light, less attention to building structure for production tables.
1 Like
@thumper knowing sex 10 days from sprout allows me to save all the time, light, nutes and attention for plants that are never going to make it to the production room, I keep the males for breeding just under VERRY dif conditions
4 Likes
@tokesandtinkery Dig it out man, that would be fun to mess with!!!
1 Like
I’m looking to get it all setup! Just gotta find a space to setup in
1 Like
Medicinal Genomics
3 Likes
Doesn’t have much utility, as it’s not accurate beyond the sample being tested. Plus, you’ll lose money in overhead costs, by having to destroy up to 100 seeds per analysis - just to get enough DNA to test for the presence of males from that single sample.
Better off ensuring the STS femminizing process is being done properly and/or do it yourself.
1 Like
That’s a calculus for someone else. I would imagine if you’re about to plant thousands upon thousands of seeds for a farm, the cost of having (completely made up numbers) 1% of plants ending up males versus say 2% probably has huge upsides for the cannabis farmer.
I don’t know the value of one male plant and its ability to ruin the crop around it. I imagine its probably more than the $20 for the test and cost of 50-100 seeds.
1 Like
The test is merely ensuring the sex of 50 to 100 seeds analyzed. If you’re only analyzing the DNA for 100 seeds, it doesn’t determine anything for the remaining 99,900 seeds. Likewise, the test doesn’t identify seeds from monecious lines either. I just can’t see the utility.
You wouldn’t test only 50-100, you’d test much more in that size batch of seeds.
Are you thinking about it from a costing perspective? For example, how much money is spent fertilizing, watering, laboring and other expenses that go into a crop before it gets damaged by surprise male plants? What happens if you’ve put all your investments into a grow and it turns out the seed supplier scammed you and sold you mixed seeds, but you’ve already gone ahead and planted them? When do you find out? 6 weeks later? 8? What is the opportunity cost of that?
Hopefully someone closer to large scale growing chimes in, it generates some interesting questions!
What does the infrastructure for 90,000 plants look like? What does that cost to setup? Land, equipment, infrastructure. What does that cost to run and operate? How hard is it to identify and remove males in such a sized grow? Is that prone to human error and if so to what degree? How do those costs compare to lost quality, yield or possibly entire crops due to males? What is the market value drop of your crop going from no seed, to lightly seeded, to heavily seeded bud?
I see tons of utility in being able to get ahead of a potentially harmful ($$$) plant showing up in my crop.
100 seeds randomly selected from a homogenous batch is representative of the entire batch of seeds (depending on size of batch). That’s how probabilities work… If none of those hundred are male, then probability is X% that the remaining batch is free of males. This is if the seeds were randomly selected only and not just from one bin or something.
Also monoecious could be tested for If the pcr was designed correctly but you’d need to make sure of that with the lab.
2 Likes
I just don’t see how if you test 100 seeds from a cultivar that can easily yield over 3,000+ seeds >per-plant< … that it’s an accurate representative of the entire batch. It just seems like the focus should be centered around finding a reputable breeder who knows how to properly mix/apply STS. I’ve never encountered a male plant from feminized seeds, but either way … the risk seems absurdly low, unless the breeder is a complete hack.
So how many sacrificed seeds would it take to accurately conclude that you would not find a male out of a 100K batch?
Fair enough. Do you know if there’s actually an issue with industrial hemp crops being ruined by males from feminized seeds? Are the hemp breeders really that terrible at doing reversals?
Nonetheless, I see more utility in finding reputable breeders who properly mix/apply the STS treatments in the first place, because virtually eliminating the chances of finding males is precisely what the feminization process does.
But Elemental Processing of Lexington, Kentucky, claims those plans tanked when HP Farms of Troutdale, Oregon sold it more than 6 million seeds that were mostly “male.” The female plants produce a CBD-rich flower.
The suit says it was only after the Kentucky company had distributed the seeds to farmers and those farmers’ crops had sprouted that the company learned the seeds were male.
Elemental Processing estimates it lost at least $44 million in profits.
One quick example, it might be the one that was in the back of my mind somewhere thinking about this issue.
Imagine all that could have been saved with this kind of inexpensive and quick testing.
Certainly, that’s good business sense. The question becomes how do you get to this point when you have none to begin with?
For me, I follow the mantra ”trust, but verify”.
1 Like
Accurate if ya stir the bag of seeds to make sure the 100 seeds didn’t all come from one individual plant… Accurate as in providing an estimate of <1% males for said bag of seeds… Test 200 and none are males… Now you are estimating <0.5%. Course don’t let them brazil nuts float to the top and screw your sampling hyup.
If the seeds were produced at the same time and in the same way - you would expect a normal distribution. If you had a normal distribution then you could do an ANSI/AQL sampling of the seeds and test those. And then you would have +95% confidence that you were correct.
Which is a lot better than doing nothing at all. And is a lot better than having to test every plant after. And is certainly good if you are receiving a batch of seeds from a new supplier and you want to test to see if their projection of the percentage is even remotely close to reality.
Its all about mitigating risks through statistical destructive testing. Just like not every single bullet is tested from a batch produced - but instead only a handful of the batch.
3 Likes