A Stab at explaining Mendel (breeding)

Correct Racer, that's why a good strategy is to stabilize two different strains, and then cross them for the hybrid vigor. F1 hybrids also are very similar to each other. If one is AA and the other aa, then all the F1s will all be Aa for that triat, since they have to take one from mom and one from dad. this is part of why F1s are desirable for growing. For breeding on the other hand, AA or aa is preferable, because the results are easier to manipulate.

if cloning you can sometimes find the winner in the first batch of f-1's you have made. when i do i cross to a known winner of clone only genetics in which case someone has also picked a favorite for mainstream use. it is in a way cheating but my space is limited so if i can use a strain already tested for its general attributes and stability i will some great strains can be produced this way in my experience.

bicycle racer said:

if cloning you can sometimes find the winner in the first batch of f-1's you have made. when i do i cross to a known winner of clone only genetics in which case someone has also picked a favorite for mainstream use. it is in a way cheating but my space is limited so if i can use a strain already tested for its general attributes and stability i will some great strains can be produced this way in my experience.

Click to expand...

theres no cheating in breeding my brother..the early bird gets the worm...yeah a real easy way to make a strain is to get two known genetics of the same type (I.E indica or satvia) that are ture breeding and stalbe and cross them in a IBL program....MEANGREEN.

wait what???????

A 'winner' is merely a good candidate for breeding.. It carries no promises since A+B <> (A+B).. I agree there is no cheating in breeding, but cloning does allow you to work with the past in a way, and hind-sight is 20/20..
I stress again that the cloning only allows you continued access to a good 'candidate' for hybridization.. Results from the cross may not be what you expected at all, or even more likely they won't be anything near uniform since it takes two damn true breeding parents to acheive that..
That being said, hybrid vigor is real, and its quite likely you end up with an assortment of strong phenos simply by crossing two good parents regardless of what you know about their genotypes..

The problem is that breeding isn't as easy as Mendelian genetics. More often than not genes are linked. In a Mendelian system each trait is controlled by one gene and the genes are independent, segregation of one from another. Usually traits are linked to several different genes, each gene can control more than one trait, and there can be physical linkage. Then recombinance throws a wrench in.

Simply put if you crossed a plant dominant purple recessive low thc with a plant recessive green and dominant high thc; you wouldn't necessarily get a purple high thc plant. And this is a real basic down and dirty example. Breeding class makes the brain hurt.

LightGrower said:

The problem is that breeding isn't as easy as Mendelian genetics... .

Click to expand...

Oh true enough. But look at a wolf, then look at a chihuahua, and tell me again it's too hard for amatures to make progress to bother with.

Simply put if you crossed a plant dominant purple recessive low thc with a plant recessive green and dominant high thc; you wouldn't necessarily get a purple high thc plant.

Ok, so you are decribing
P?tt x ppT? where P is color and T is thc

So we expect to get

...........Pt..........Pt..........?t..........?t
pT...pPTt........pPTt......p?Tt........p?Tt
p?...pP?t........pP?t......p??t........p??t
pT...pPTt........pPTt......p?Tt........p?Tt
p?...pP?t........pP?t......p??t........p??t

Out of the sixteen possibilities, we are only interested in PpTt (4) for sure, and maybe more depending on the values for ?. We might get lucky or not.

So we know that about 25% will likely be what we are looking for, and plant enough to have fair odds of it showing.

You would know you shouldn't have to grow thousands of plants to find the combination once.

Also keep in mind that it's the number of seeds used from a single generation, you can do 4 grows of 25 and get the same results as if you grew 100 at the same time.

If I am off in my calculation, I'm buzzed, post a correction for everyone else's benifit, and get over it.

Are there exceptions? You bet. Does it require luck? Yuppers. Are some combinations impossible? Likely. But I refuse to buy that the only people who can do this is some secret society of corporate breeders who use magicial powers to alter the genetics. Well, until they start gene splicing and such.

I mean absolutley no disrespect towards commercial breeders. that's a job I'd do a donkey for.

You need a large selection because you can't 'really' boil it down to those permutations..
You might have (P1P2)?(t1t2)x(p1p2)(T1T2)? or any scarily complex recipe where traits are controlled by many unlocked alleles.. Sure there are broader Punnet squares, but no matter how deep you go, you'll never truly know the genotypes, and the more plants you'll need to observe to spot inconsistencies in your model..

born2killspam said:

You need a large selection because you can't 'really' boil it down to those permutations..
You might have (P1P2)?(t1t2)x(p1p2)(T1T2)? or any scarily complex recipe where traits are controlled by many unlocked alleles.. Sure there are broader Punnet squares, but no matter how deep you go, you'll never truly know the genotypes, and the more plants you'll need to observe to spot inconsistencies in your model..

Click to expand...

You are correct, but that is no reason not to try, and Punnet squares will give you more informed guesses than just randomly mating plants and not having any clue how to interpret the results.

Simply breeding your best to your best will generally give better results than just mating any two random plants in your collection.

Some people like to just buy good seeds and then take clones from the seeds, that's cool. I'm just putting in my 2 cents for those interested in mucking about with breeding.

It just seems that you're more bound/determined to make the results meet the model than the other way around.. Breeding, and the scientific method really go hand in hand..

Naw, I view it a lot like playing poker, the better you understand your odds and the way the game is played, the better off you are in the long run.

As long as you remember that there are more correct/complicated methods of calculating odds than the simple calculations ppl use at the table..

born2killspam said:

As long as you remember that there are more correct/complicated methods of calculating odds than the simple calculations ppl use at the table..

Click to expand...

So spread the wealth, enlighten us.

Most ppl only calculate the odds that they hit a hand using the rule of 4 & 2.. That doesn't necessarily correlate to winning the hand for instance.. Just an example, there are many others depending on the situation at hand..

born2killspam said:

Just an example, there are many others depending on the situation at hand..

Click to expand...

I meant do you have anything to contribute about how to wind up with cool plants.

As far as breeding goes, I know the theory.. The theory itself is pretty simple to me since I've probably spent a couple thousand hours grinding through statistical physics..
What isn't simple is the feasibility of accurately determining genotype of a hybrid strain by assuming a simple Punnet square, and the number of plants needed to test a more complex square grows exponentially..
The Punnet square is more feasible if you define your properties on the basis of your F2s or later, after the initial crosses.. That way you'll have cancelled out alot of codominance factors, and simply accepted the codominant property, and observed new properties unlike either P1.. Hybridizing opens up a Pandora's box of recessive genes that have been pummeled down within its own IBL, but become dominant when put with others.. The closer you want to get to picking/choosing combinations of traits pure to the P1's, the more selection you'll need.. The more you inbreed your cross before defining your Punnet square, the better your chances of locking down properties you've defined in your square with a smaller number of plants..

This really brings me back to sophomore biology, but it's keeping my attention better this time through cuz we're applying it to weed xD

it all comes down to crossing and breeding IN REAL LIFE! u can try and say this will happen and that will happen...but unless you REALLY DO IT..it dont mean shit! plants can do some crazy stuff.nothing is for sure on paper/screen..bottom line: just use the math with some REAL LIFE crossing find out what works and share....peace....MEANGREEN.

So does anyone know the dominate ressessive for any trait.

I've read on boards that purple color and autoflowering are both recessive, but clue if that is accurate or not.