*EDITED* hibe's RRS thread from CW!

[Telepod]

Warning, this is long. Do not attempt to read without making suitable preparations! You will need: A good attention span; a fully packed bowl; suitable food/drink; an open mind. I don't know what to do other that to copy and paste the whole schmear, so here ya go:

|^> |^> <^
|`\ .|`\ .._)

This is an excerpt of Hibe's RRS thread on Cannabisworld; it is edited for informational/teaching content only, and all deletions and inclusions were my arbitrary decision. Mostly I did not take the extra effort to preserve the identity of the user for each given post; I thought this was secondary to the informational content presented, and I wanted to stay away from the politics that had diluted the thread to begin with. So, here is a lot of useful reading for anyone interested in genetics and breeding theory... enjoy!

-Telepod





Reciprocal Recurrent Selection(RRS)-modified for cannabis.
(paraphrased by hyb.Originally paraphrased by Chief Seven Turtles)
RRS is a very applicable breeding procedure for cannabis,and used to maintain beneficial genetic diversity, and to exploit hybrid vigor. The strategy uses hybrid testcrossings to uncover the genotypes that will breed most favorably,and create the Ultimate F1 seedlot.
This model should enlighten you to OTHER options and strategies that keep this outcrosser diverse and heterotic. Keep in mind, RRS is not the only way to reach the goals we have for cannabis. I hope this gift to the community can educate,and create a new and better direction for cannabis breedings.


RRS: The first half-cycle:

You begin with 2 distinct varieties. Diverse populations give rise to the most genetic GAIN-we'll call them Populations A and B.This Original seed generation is labeled "Gen0(zero)"..The A population is labeled A0, and the B population is labeled B0.
Grow and flower as many plants as possible of A0 and B0, and keep them completely separated from each other.This requires two separate grow environments.
You have TWO pollen collections to do, and FOUR crosses to make with them, as follows:

1)MIXED male pollen from the BEST B males is collected and testcrossed to each single,ideal A female of your choice.(as many as possible)

2)The SAME Mixed B pollen is also INBRED to the best Bfemales youve selected.

3)Mixed male pollen from the best A males is collected and testcrossed to the selected B females.

4)Mixed A pollen is also INBRED to the SAME ideal A females that were crossed in item 1 above)

.......So, each population's select females have been pollinated by a mix of its OWN FAMILY'S POLLEN(the inbreed),AND a mix of the other strain's pollen.(the outcross or crossbreed)...AND LABELED!

Harvest the seed from all these inbreeds and outcrossings,and keep them SEPARATED and labled accordingly.

The second half-cycle:

Grow and flower ONLY the CROSS-BRED LINES(A#2,4,8.. x MixedB,etc..)- NOT the INBREED LINES. *just hang on to those seeds right now.We will use them later....*

Evaluate and determine which crossbreedings were the most beneficial to your goals.Goals that include potency, obviously will require an evaluation by smoking the testcross females first.EVERY GOAL IS DIFFERENT,and you will have to decide what yours is,and modify your strategies accordingly.

Lets say that during your progeny test, you find that 'mixed A pollen' gave the best overall progeny with B females#1,5,7,15,35,and 45. We will say for now that 'mixed B pollen' bred most favorably with A females#2,4,8,14,36.etc.

CONSOLIDATE all the seeds from the most favorable INBRED lines of A females (A#2,4,8,14,36etc x AMixmales) into ONE container and label them Seedlot A1. Combine/consolidate the INBRED seed from all the most favorable lines of B females (B#1,5,7,15,35,45etc x BMixmales) into one container and label them Seedlot B1.
This new generation of INBREED seed(A1 and B1) has now been verified that its various genotypes give the most favorable progeny when crossed into the other distinct genepool.
They will not actually be crossed together until several(or more) generations of RRS..You have just completed ONE generation of RRS. At this time,DISCARD all the remaining seeds of the non-selected INBREEDS and OUTCROSSES.


BEGIN YOUR SECOND CYCLE OF RRS.

Again...grow out and flower your new A1 and B1 populations in separate environments.
Collect mixed pollens of populations A1 and B1.
Cross pollinate AND Inbreed the ideal females of each population.label.
Grow those testcrosses out and determine which lines give the best females of each and again,
Keep ONLY the seeds from the INBRED female breedings that have proven to be ideal through TESTCROSSING. Combine these seeds into your A1 and B1 containers..(they now contain 2 sets of RRS each,but have never REALLY been intercrossed,see??? They are the inbreedings of the most favorable outcrossings.)

...continue Reciprocal Recurrent Selection until you feel you have reached the highest degree of combining potential these 2 populations can offer.

*Finally, Grow out a sample of both RRS'ed seed POPULATIONS and use one population as the pollen donor and one as the egg donor.Again, culling anything that doesnt conform to your project goals.You have just created a REAL F1.


can you see any differences from the ones youve been buying?

Any questions?_____________________________________________

Id like to thank Chief Seven Turtles for his rendition of the RRS strategy.

___________

Hyb, the method you mentioned will work and has some good points but it has some drawbacks too. For one thing, as you pointed out, selecting for one quality will tend to let other qualities diminish. You said nothing about maintaining disease resistance or selecting for multiple traits. It can be done. In doing the type of project you outlined you aren't going to be selecting for specific genes but for traits or qualities. As I said before, a trait will involve many genes and unless you are using cutting edge laboratory technology involving gene splicing you aren't going to select for genes. What you are talking about is a refinement of ordinary breeding technique. As Tim already said, it's not cost effective to do it that way and you can get similar results using less time consuming methods.

Another important point is that you did no selection at all on the male side in your write up. The male contribution is just as important as the female's. Let me suggest another way to do it.

Take your A and B population and grow them out separately. Take clones of each plant and flower the clones but keep the original plants in veg. From the flowered clones take pollen from each A male and pollenate a branch of each of the best B females. Likewise you can take pollen from B males and pollenate branches of A females. This can be done reliably without risking accidental cross pollenation. Put the pollen samples into plastic bags and seal around the branch being used. After each branch has it's bag of pollen let them set for a day. Remove each bag one by one and when a bag has been removed spray the branch thoroughly with soapy water. Cover again the branch with a plastic bag and remove the bag from the next branch and spray it. Repeat until each branch has been pollenated, sprayed and covered. After you have done all the branches remove all the plastic bags and let dry. The water zaps the residual pollen and prevents it from getting on other branches.

Now you have your prime females each pollenated by several different males but you can label the branches and tell which was pollenated by which. Finish the seeds, plant them and grow out the crops only flowering the females. By growing several females from each test cross you can see which males are the best and which females are the best. You also see which males do best with which females.

You now go back and you know which of the original males and females are the ones you want to use with your breeding program. Remember you have been keeping them in veg state. Keep them in veg and take more clones from the ones you want. These will be your breeders. Discard the males and females that didn't work out and you have your mother plants that you can work with for years.

The advantage of doing it this way is you only need one test generation and then you can start breeding. It also saves on grow space but if you have the room use a whole female with each male you're testing. The problem with that is you need a lot more space. There is nothing wrong with Hyb's method but it is much more expensive and time consuming.

_______________

JW:"What you are talking about is a refinement of ordinary breeding technique. As Tim already said, it's not cost effective to do it that way and you can get similar results using less time consuming methods."

>>>A refinement?ordinary breeding technique? i dont understand what you mean. RRS is only ONE application a breeder can use to improve plants.In fact,some breeder who has certain extenuating goals, may only use RRS as a mere STEP in his overall program.

What you describe is TESTCROSSING, and determining Specific Combining Abilities between single genotypes,not the SCA of a population,like RRS does.Your model is not much different than RRS,except that your model requires far more testcrosses and progeny tests to get to where RRS takes you.
If you were to choose 5 males and 5 females to do your first round of testcrosses(your model), you will have created and will be required to grow out 25 LINES of progeny to determine the most favorable (single-genotype x single genotype) matings.(and i hope you didnt think you were finished in one cycle of recombination,unless you were planning on growing thousands of each of your 25 testcross lines and THEN selecting improvements from them?)
RRS does more work for the breeder by selecting a Sample of the opposing GENEPOOL with the mixed pollen. RRS takes a sample of the genetic options from each pool and tests it to multiple single genotypes,in order to increase the frequencies/occurance of required genes in the 'working genepool', for the ultimate purpose of increasing the PROBABILITY that you will fix a desirable genotype in the future. By using the mixed pollen, you increase genetic diversity,and this is another IMPORTANT goal of a consientious breeder. As well, the amount of progeny required is far less per growout than your method of merely making many testcrosses in one pass.

You will discover that there is no 'miracle shortcut'. If your program requires 25 testcross lines of 25 plants per generation(625plants),then another program may require 5 CYCLES of 5 testcrossings of 25 plants(625total,125plants per) to get to the same degree of genetic gain.RRS IS A SHORTCUT.


JW:"Another important point is that you did no selection at all on the male side in your write up. The male contribution is just as important as the female's."

>>>>Thats not true. The breeder's choice of which males contribute to the mix of pollen was certainly 'selection'. You dont have to tell me how important the other half of a diploid's genome is.


JW3quotes)"Take your A and B population and grow them out separately. Take clones of each plant and flower the clones but keep the original plants in veg. From the flowered clones take pollen from each A male and pollenate a branch of each of the best B females. Likewise you can take pollen from B males and pollenate branches of A females."

"....Now you have your prime females each pollenated by several different males but you can label the branches and tell which was pollenated by which. Finish the seeds, plant them and grow out the crops only flowering the females. By growing several females from each test cross you can see which males are the best and which females are the best. You also see which males do best with which females."
"You now go back and you know which of the original males and females are the ones you want to use with your breeding program. Remember you have been keeping them in veg state. Keep them in veg and take more clones from the ones you want. These will be your breeders. Discard the males and females that didn't work out and you have your mother plants that you can work with for years."

>>>>>>You have just:
1)TRIPLED the number of pot plants in your possession, to do ONE CYCLE of selection.(Originals+clones you flowered and pollinated+Original plants to now breed with)..EXTREMELY not good.
2)You have introduced cloning into a strategy that doesnt require multiplication,thereby introducing more space requirements and the increased possibility of sample error from failed clones.
3)Wasted a lot of time.You could have done at least 2 cycles of selection in all this time.
4)You have caused yourself to need to do 2 complete flowering cycles instead of one,to get your ONE cycle of selection done. More risks unnecessarily introduced by your method.
5)You have created far more testcrosses and therefore,far more progeny growouts than necessary.This leaves the breeder to select from smaller samples,which increases error and decreases efficacy.
6)After allthis cloning,labeling,growing,flowering,recloning,flowering stuff,you still have only done ONE generation of selection..you have not even accounted for the benefits of 'recombination', found in doing more cycles of selection with increased genetic diversity.

RRS is the shortcut.Based on your method, RRS is a more accurate method, far better suited to the heterogenic nature of cannabis(diversity),less time consuming,and requires less space and less plants. I dont recommend it....

________________________

Since this forum has overall shown much more maturity, I'll offer my own challanges. I really like the dog model because it comes closer to reflecting cannabis breeding than most plant breeding models. This is because both dogs and drug cannabis are seperate sexed and therefore an individual cannot naturally breed with itself. Most plant breeding models are based upon plants that can self fertilize and are therefore much more tolerant of inbreeding than dogs and cannabis. Another parallel is that dog breeders work with relatively small populations, a reality for most cannabis breeders, but not one for most other plant breeders.

Now when we look at the domestic dog genepool, we see two relatively distinct types of genepools, the inbred pure lines and then the hybrids (mongrels). What is interesting is when you look at the genetic health of each group, the mongrels seem to be quite healthy, but several of the pure breeds are in serious trouble. What's ironic is that most pure lines are the result of focussed breeding efforts where the breeder was trying to improve the genetics and most mongels are the result of very little in the way of artificial selection. Mind you, many professional dog breeders don't understand genetics any more than many on this site. This tends to suggest that nature is better at picking the best genes than we are. At least when we are not fully educated on the implications of selective pressure.

Another point to consider is that the dog breeders are not going back to the wild stock to repair their purebreds, they keep working from within the lines. Many have suggested that cannabis land races are the saviour of the drug cannabis genepool, but if you really consider it you will find that option is about as useful as using a wolf to repair the poodle genepool. The concept of using landraces is from plant breeding models where the plant being fixed is much more tolerant of heavy inbreeding than pure outcrossers such as dogs and cannabis. It may be a valuable option for wheat, but not nearly as much so for highly refined drug cultivars. Incorporating a land race into a highly refined drug cultivar would be a step backwards in most cases. Plus there is no guarantee that the landraces will always be available in their true form, we are better off planning on not having access to them in the future and acting accordingly.

Now at first glance, one could argue, and I have myself, that the mongrel cannabis genepool should be as immune to improper breeding as the mongrel dog genepool. I have argued that the vast pool of indoor hybrid cannabis seeds are serving as sort of a new landrace of cannabis, the indoor landrace, full of genetic diversity. I in the past have argued that such an indoor genepool would naturally preserve the indoor genepool's genetic diversity simply by the virtue of the diversity of breeders involved. I argued that it was our insurance against the heavy inbreeding performed by the serious, although not always competent, breeders.

However, there are some subtle and important differences. The mongrel cannabis genepool being offered by incompetent breeders are not the result of basically natural selection as is the case with most mongrel dogs, but artificial selection directed by us seedmakers. And selection is where genes are fixed or lost. Selection decisions are made by every cannabis seed maker and usually by ones that don't fully understand the implications of their selection decisions. Therefore, because of the combined artificial selective pressure, uneducated people performing the selections, and the individually small population sizes, one can expect the genepool to diminish over time. Each time a fad is adopted by the seedmakers, it will effect which genes are removed and which are fixed. Who knows if breeding for frosty nugs is also fixing vulnerability to powdery mildew? We don't, but most uneducated seedmakers will select for frotyness without giving it a second thought. And just what is frostiness, does it mean more potency? My research suggests it doesn't, that it just means that the trichomes have longer stalks. Is this of significance to growing the plant outside? To shelf life? I've never read a seed maker or grower seriously address this question. But the fad of selecting for frosty nugs exists and most closet breeders are also selecting the 1000s of other genes that are linked to the frosty nug genes without giving it a second thought. It's not hard to see how a large pool of breeders can unintentionally work together to push a genepool in a direction, and not always a favourable one. Just like how dog breeders have collectively bred for hip displacia without any of them intending to do so. A trait that most dog breeders find favourable is linked to the hip displacia gene and is being selected for. Most of us making seeds do not understand the effeects of selective pressure well enough to not take Hyb's words seriously. He is offering ideas on ways to reduce the unintentional side effects of our selective pressures.


Everytime we pollinate a plant, we should be asking ourselves what are ALL the genes that are coming along for the ride. Unfortunately, very few know how to answer the question. And most don't care.


------------------
**************************************
BC Growers Association

_______________
Irish wrote:
>hey joey,
i see. i thought he was saying to label and selectively pollenate. but i see where you are saying he has "mixed male pollen."

>hmm. yeah that would make it tougher to decide.

>so i guess we are back to clarifying "how is the best male pollen picked for the mix pollen?"

>stature? smell? smoke test? progeny?

>in a mixed pollen test you will find which females respond best with the mixed pollen, and create the desired offspring, but not which male produced that pollen, because at this point you are still generalizing the breeding (i.e. inXes and outXes, but not specifics.)

>do i have it right so far?

Yes, you do. I'm not saying his system won't work or is a step backward. I'm saying it's a good system but it has some weak points which can be fixed or improved upon. The main glaring flaw is mixing multiple male pollen together. You would be better off just picking one and using only it. At least you would then have consistancy in your results. Better still is to label and test each male separately and record the results. You then test out and see which male gives you the results you want.

Lets use the example of growing for potency. It's a simple factor and *most* if not all people will agree on potency. I know there are various types of potency like thc/cbd ratios, the indica vs sativa high and so on. Lets say you are looking for the stoniest sativa high you can get from a strain. You take your 5 males and cross them with your 5 females. That gives you 25 different types. You then grow them out and after taste testing you decide that females #3 and #4 are the most potent and males #2 and #5 have the stoniest offspring. After further taste testing you decide male 5 x female 3 gives the best of the lot. You then go back to your plants that you've kept in veg and crank out a bunch of clones of female3 and some of male5. You make seeds from that combination and only that combination. As Vic said, you could lose some other qualities at the same time but you would be maximizing potency.

In answer to your question of how to test the males, you test through their progeny. IOW, you test the potency (or other factors) of the female plants grown from seeds made by pollen of the male being tested. This points out the need to use cloning and keep the original plants in veg. Otherwise you can't go back to the same combination you just got done testing.

Vic, I'm glad to see you in on this discussion. You said:

>Joey, you may not like the tone of Hyb's message or even the message itself, but it doesn't make it any less valid. If you make seeds, sucking up your emotional ego and concentrating on his message can only help you to become a better seed maker. You have nothing to loose but your pride, pretty small pricetag if you think about it. Glad to see you care enough to be challanging the specifics

I welcome a frank discussion of mj breeding by people who have a thorough knowledge of it. I'm trying to ignore some of the tone in favor of the information being given. As I've said, I do think what he posted is a good system, I just think it can be improved upon. You yourself have spoken about testing and selecting males and the importance of that. If one of the males in the pool had genes that would give unacceptable stretchiness, for example, wouldn't you want to eliminate that? By mixing all the pollen together all you would know is that some of the seeds produced very stretchy females and you would have no way of eliminating that trait directly. Same with good qualities, if you only selected on the female side you wouldnt know how to isolate an extra good trait coming from only one male. The best you could do is capture it indirectly by picking the plants that show it and breeding with them in the next generation. Sometimes you have to do it that way but by selecting males you do it directly.

I welcome your observations on the comparison between breeding of other species to the breeding of cannabis. I think there are lessons to be learned from that although one can't be too literal in the comparisons. The selecting for "frostyness" is a good example. That's just a cosmetic feature and not really important. What we have here is the power of the consumer. He or she wants pretty looking plants to impress friends so pressure is put on the breeder to provide them. White Widow is an example of a plant that while good is popular as much for cosmetic reasons as for it's overall quality. Important traits could be left by the wayside in the pursuit of less important ones. I would suggest that blue or purple coloring might be another example of that.

Vic, I do hope you will stop in more often and educate us when appropriate. You don't see the petty flaming here that you do in some forums. I for one am serious about the subject and I'm always eager to learn.

Joey Weed

______________________

Joey Weed,
The reason you are using your top two thirds or so males is to keep the gene pool sufficiently wide. Selecting the single best male to go with your single best female narrows it all down to a "foundering" event, in case you bothered to read the original dog breeding case history, link at OG. I encourage you to read and understand what it is saying when it comes to foundering events and bottlenecks.

After using all the top males, obvious seedlots which are more vigorus and desirable in the end will become apparent from the "f1". You will then go back to the corresponding inbred seedlot(s) and combine all your improved seeds to form the basis of the next growout (new "f1's" to select new, improved inbred seedlots, to make new "F1's" etc). Absolutely no reason to keep any clone mommas unless it is for other reasons. You can improve your crosses in this manner until you actually kill too many braincells from the improvements you have made!

_______________
Hey Joey, Hyb's presentation of the RRS concept isn't an absolute recipe, no breeding blueprint is. The idea is to learn an overall concept and then adapt it to a particular project. One breeder may mix their males as Hyb suggested and another breeder may not as you suggest. The blended pollen does not make the RRS concept, the reciprocol test crossing does. I would let my base genetics decide which route I was going take. Like say one side of my RRS project was as finished as blueberry, then I would probably mix the pollen as per Hyb's model simply because there isn't alot to be gained from selecting between blueberry males, they are all pretty uniform. Pick the top 5-10% for pollen sources and move on. However, if I was starting with an unfinished IBL project like say a cubing, then I would definately segregate the pollen from each male because I would have more selection to consider. Not only would I be looking for superior combining ability with each cross, I would also be looking for the males that were the most homozygous. Hyb's model assumes a more "finished" starting point, that's all.
Same point holds true for a cubing project, there is no absolute way to conduct a cubing project, there is the overall concept, but the breeder must allow their specific genetics and purpose of the cubing to dictate the precise methods.


------------------
**************************************
BC Growers Association
__________________

okay lets cut this back a bit...

no dogs for a second, or hell, i even have a dog anecdote...

searching for density, planning on using the RRS. buy a bunch of "high calyx/leaf ratio" beans, say two seperate kinds, 25-50 each.

you find some stocky, stinky, close noded, lightly bushy males. say ten of each strain.

collect and mix the pollen of ten "A" males.
collect and mix the pollen of ten "B" males.

make "YG" x "A" mixed seed. (YG is your girl)
make "YG" x "B" mixed seed.

at this stage clones are not necessary. of either the A or B mixes, and not the favorite YG. because we are focusing on finding the new traits in the offspring ("bred" stock), not "breeding stock".

majority of YGA (gen zero?) makes heavy fruity bushes, but fluffy.

majority of YGB (gen zero?) makes dense, leaf-barren bud, but not heavy yield.

now we make YGAf x YGAm, and YGBf x YGBm, and YGAf x YGBm, YGBf x YGAm. (males and females.)

so we are taking the proven female and seeing what her babies do, then getting her out of the equation, no back breeding? not cubing her variable traits? but solidifying her offsprings traits? to a better end.

i am just trying to clarify the starting point, if attempting with one certain clone.

please, throw in, or out, what is my major malfunction here.

i have mentioned this before when the topic has come up, but there is a guide to "aiding hybrid vigor" in clarkes book, that shows a flow chart of this, but with more of a selective process, so it would fit more along joey's style. where as hybe is not wanting or needing to know which daddy is better, he is looking at which family is better.

so i am assuming again, joey is looking for a keeper mom and dad, and hyb is laying out a keeper "family tree."

so with dogs, better to have the ***** impregnated by three or four stud males than one male to one batch. increasing variety, instilling a better genetic disposition to offspring.

the *****es traits will come out no matter who the daddy is, but you can also see the daddy of each in the litter. maybe to help you know the brown ones get displacia, or the white ones are deaf.

dammit, it sounds too simple.
LOL.
__________________

Irish, no mom clones need be kept as you are gathering excellent alleles into two seedgroups that have optimum combining ability. You will end up with final F1 seed batches where every female is likely keeper quality.

Four ten packs of diverse beans are better to start with than two 10 packs of uniform beans. You want all the diversity possible at the outset because you will be honing how they all fit together in the later cycles of breeding. Does that make any sense to ya?
___________________

LadyG wrote:
>The reason you are using your top two thirds or so males is to keep the gene pool sufficiently wide

That is one way to do it. If you are looking for variety then by all means use as many males and females as possible. The thread was about selecting for desireable traits and qualities in a seedline or in a cross of two lines. In that case you don't want variety you want a certain trait. It really depends on your goals what method you should use.

Vic High wrote:

>Hey Joey, Hyb's presentation of the RRS concept isn't an absolute recipe, no breeding blueprint is. The idea is to learn an overall concept and then adapt it to a particular project. One breeder may mix their males as Hyb suggested and another breeder may not as you suggest.

Sure

>The blended pollen does not make the RRS concept, the reciprocol test crossing does. I would let my base genetics decide which route I was going take. Like say one side of my RRS project was as finished as blueberry, then I would probably mix the pollen as per Hyb's model simply because there isn't alot to be gained from selecting between blueberry males, they are all pretty uniform

Good point, the more uniform the material you are working with the less need for selection.

>if I was starting with an unfinished IBL project like say a cubing, then I would definately segregate the pollen from each male because I would have more selection to consider. Not only would I be looking for superior combining ability with each cross, I would also be looking for the males that were the most homozygous. Hyb's model assumes a more "finished" starting point, that's all.

>Same point holds true for a cubing project, there is no absolute way to conduct a cubing project, there is the overall concept, but the breeder must allow their specific genetics and purpose of the cubing to dictate the precise methods.

OK, no argument. It's just that Hyb started out by talking about how to improve by selecting for a trait. If you are aiming for a certain trait, like potency for example, you will want the advantage of selecting on both sides. This assumes there is enough variability within the strain or within the two strains you are using so that there would be some difference between males. Of course if it is uniform to a certain degree then there would be little point to any selection at all. You could use any healthy male and female to make seeds and do alright. In order to benefit from selection there has to be some degree of variability in the parents and offspring. If the only observed variation was, lets say, two phenotypes then you could select from the phenotypes to see which one was best for your purposes. Then again you might want to combine the two rather than segregate them in order to get benefits from each.

Irish, I see what you're saying. You can look for keeper moms and dads or keeper familys. If you have a huge amount of variation in a strain you will want to narrow your search for the keeper mom and dad, I think. It would depend on what your end goal was. If you want a seedline that is uniform and has certain characteristics then you don't want a huge amount of variation and you will try to focus on just what you're looking for. If you goal is to combine as many different qualities together as you can then you would work more with groups than individuals. Most seed buyers want a uniform strain. They want a lot of super females in each crop. They don't want some that are tall, some short, some skimpy and some heavy producers. They would like uniform height and uniform heavy yield with high potency and good taste/smell. This may not be the best thing for the gene pool but if the breeder doesn't cater to the buying public he will not be successful.

What Hyb said is valid and I recognise the need to maintain LRG's and keep some variation in the seed stock so that they don't go downhill. Still and all, the breeder is sometimes like a horse trainer working with wild or semi-wild animals. The buyer wants an animal that's tame and follows orders. The breeder then has to *break* the horses to get them to the specifications of the buyer. The breaking process would in the plant world, involve eliminating excessive variation and selecting for outstanding and desireable qualities. In the process some genetic variation is lost and with it some of the depth of the gene pool. Perhaps we need to maintain wild and semi wild stocks in their pristine form so that we can draw from them later on?

Joey Weed
___________________________

>>>>>If you are aiming for a certain trait, like potency for example, you will want the advantage of selecting on both sides.

Not necessarily true when you are addressing polygenic traits such as yield or potency. These traits tend to not be highly heritable traits, meaning that they are often hard to select for on an individual level. If a breeder wants to improve these types of traits, then the breeder must select families and not individuals to get the greatest genetic gain for his/her efforts.

Another part of your argument keeps suggesting that Hyb's model doesn't offer any selective pressure on the pollen donor, but this isn't true. The pollen donors have already undergone a round of selection prior to each reciprocol cross and the cross is only made up of the selected individuals. All that happens is that the pollen donors are not selected as to their combining ability with the seed parent. But since this is done with the seed parent, adding the extra work on the pollen side could be considered redundant since you have already selected the best seed parent for a pollen family. And done the visa versa on the reciprocal side.


------------------
**************************************
BC Growers Association

_________________

The whole idea of using multiple males in RRS, IS to maintain diversity -for the purpose of adaptability ,health,and Gain et al.- while you SLOWLY increase the frequencies of the favorably-combining genes.Less-severe pressure provides more diversity,and hence, more adaptability,health,etc...its a balancing act...

We need to remember several things:
Quantitative traits like potency,height,whiteness,aromas/flavors..can require
a 'cumulative effort' from multiple loci,to make the trait appear the way we want it.These alleles involved, NEED to become homozygous at the loci,or at least be in very high frequency at the loci if we want to achieve 'uniformity' in our F1.Both methods can do that,although Joeys method does not afford the population enough chances for improved genotypes, as a result of the diversity found in mixed male gametes.
By maintaining the allelic DIVERSITY held in the MIXED pollens, the breeder is able to make more Gain by selecting from RECOMBINATIONS of this diverse pool of alleles, in the progeny tests.Aswell,More males in the RRS pollenmix does not increase the number of testcrosses.Unlike Joeys method.

RRS is designed to slowly increase the gene frequencies at those multiple loci controlling the desirable traits, so that the POPULATION, WHEN CROSSED will be uniform. With each cycle of RRS,a 'slow but sure effect' occurs.
This broadened,and slower process of genefrequency shifting in RRS--(remember,in tandem with Environmental pressures and RECOMBINATION from crossover events at meiosis,so therefore,more interventions/interactions occuring),- permits more diversity while still allowing for the breeder's DIRECT and INDIRECT selection pressures to allow him to surely reach the goal. Joeys one x one plant testcrosses are in no way as 'encompassing', in terms of the potential for favorable recombinants resulting from the sheer numbers of test matings possible in mixed pollen.RRS male numbers can increase,without any more numbers of testcrosses involved.
Statistically,as we know, the odds of improved genotypes are better in higher numbers.As well, RRS is designed to statistically improve the Population variables,not just discover a few SPECIFIC desirable single-plant matings. We need to begin thinking in terms of populations,not single plants.We need to think in terms of gene frequencies,not just hetero/homozygosity.


RRS leaves the breeder with 2 separate 'improved' SEEDLOTS that can be started ANY TIME and still give the SAME,IMPROVED F1 RESULT.Joeys method requires keeping MANY PARENTS-males and females- in his possession at all times,and is ONLY ABLE to RE-MAKE the SAME (not any more improved)F1. EACH subsequent cycle of RRS performed however, will increase the potential for EVEN MORE IMPROVED combining ability between the 2 populations, with NO mother/father plants kept for breeding..With Joeys method,you would have to go through all that growing,doubling,flowering,cloning and extra time and introduced chances of error, to get back where RRS has LEFT YOU from its LAST cycle. Joey would need to start more Original population seeds,or keep many live plants around,only to continue to 'collect' more and more mother/fathers,and STILL only have subjected his populations to ONE CYCLE OF RECOMBINATION and SELECTION.

RRS keeps the two populations UNhybridized.
Joeys method selects the best matings and then hybridizes.Since he has NOT done an additional INBREED matching the OUTCROSS as in RRS,his F1 is hybridized after one cycle of selection.There is no more potential from these parents that he has kept as 'breeders' to IMPROVE the combining ability of the 2 populations anymore.Subsequent inbreedings of Joeys project result in higher inbreeding depression,due to the limits he had to place on diversity,from all those necessary testcrosses and backup clones. RRS is sitting there waiting for you to take it ANOTHER NOTCH up after each progeny test.

RRS is more oriented to use diverse starting genepools(this includes Landraces,and synthetic cultivars,multicross hybrids..mixed bagseed even..etc..),so that they will be,by the nature of RRS, formulated FOR EACH OTHER.


Does this clear it up any?


As Joey said(and it made me very happy to see that)...We cannot improve non-diverse genepools...There can be no genetic Gain when there is no potential for Gain,with no diversity. This is where one x one plant breedings by amateurs takes the genepool.


Im glad to see some thoughts..seriously,i am.

havagoodone
hibe
____________________________________________________

****LOOK****
Reciprocal
Recurrent Selection
by Chief Seven Turtles

Although many breeding strategies can be used to improve plants, I'd like to outline a strategy that would be particularly suited to Cannabis. "Reciprocal Recurrent Selection" (RRS) is a strategy in which two different populations are maintained separately but improved so that a better "hybrid" population is created when the two populations are crossed. This is done by evaluating and selecting plants within each population that produce superior progeny when crossed with pollen from the opposite population. RRS has been used to improve corn populations that will eventually be used in crosses with each other. The RRS system that I'm proposing for Cannabis is slightly different but should accomplish the same goal. Assuming that RRS is conducted properly, the following advantages should be realized:

• The hybrid population will improve with each cycle of selection.

• Although each selection cycle takes two growing seasons, the breeder will have quality herb to market after each season.

• Selection can continue until the breeder satisfies his objectives or destroys too many brain cells.

The following limitations should also be realized:

• The populations to be hybridized should flower simultaneously so that crosses can be made between them. Selection itself may converge the maturity of the two populations.

• The breeder should have access to a separate location for each distinct population to prevent pollen contamination and maintain the identity of each population. Each location should have enough room to grow at least 50 plants if possible. This will conserve genetic diversity and minimize inbreeding within each population.

Establishing goals and choosing the appropriate germplasm are essential to a successful breeding program. Assuming that you will work with only two populations, choose two that have good qualities but differ enough so that they will complement and hopefully display hybrid vigor when crossed.

Let's assume that plants in population "A" tend to have compact structure and dense flower set but average potency, while plants from population "B" tend to be lanky but very potent. We can obviously select for the ideal" compact and potent plants within each population but are limited by the amount of genetic diversity within those populations. By crossing the two populations (with controlled pollinations), we should be able to expand the genetic diversity. This will increase our chances of finding ideal plants and allowus to take advantage of "hybrid vigor" (see Marijuanna Botany, " pp. 69-71). Although we may see some hybrid vigor without selection, RRS will allow us to optimize the expression of hybrid vigor.

The figure on page 36 shows a diagram of one complete cycle of RRS. "A" refers to population A before selection. "A" refers to population A after one cycle of selection, "A after two cycles of selection, and so on. Although the figure only shows eight females from each population, at least 10 to 20 vigorous looking females should be selected for "toperosses (TCs) with the other population. To make a topcross of B onto an A female, collect a random sample of pollen from population B males (by shaking some pollen for each male plant into a bag and then mixing) and dust a receptive branch on each female in population A. Read chapter 2 of bMaryuana Botany" for a detailed account of pollination techniques. Make sure that branches to receive controlled toperosses are bagged before and after pollination to prevent pollen contamination . if population A and B are planted in separate locations, crosses of A females with random pollen of A males can be allowed to occur naturally and without bagging (likewise with population B).

In Season 2, plant a sample of seed from each topeross progeny or ~topeross family" (at least five to l 0 plants of each family and at least 20 families, l 0 from females of population A and l 0 from females of population B). This means that at least lO0 to 200 plants should be evaluated in Season 2. Some families can be eliminated based on seedling vigor if space is limited. Topeross families are then used to evaluate the female plant they came from but are not saved as a seed source. For this reason, pollination control is not necessary in Season 2.

Decide which of the A and B females are superior based on the "average" performance of each topeross family, but do not save seed from the topeross families. Instead, go back to the remnant Season 1 seed from the selected A females—seed that resulted from random matings with A males. Now mix together the remnant seed from only the selected A females. This could be seed from 10 to 50 percent of the A females evaluated depending on how many A females you have evaluated.

In this way, you will not mix genes from population A with genes from B but will concentrate A genes that do well when hybridized with genes from B. Do the same with seed from selected B females. The newly reconstituted populations will now be referred to as A and B respectively, meaning that one cycle of selection has occurred.

It is possible to improve the rate of progress from RRS by using a slightly different strategy, but this would require many more controlled pollinations. The procedure presented here requires minimal controlled crosses and can be conducted rather easily if enough space is available to evaluate 100 to 200 plants.

Remember that the final product of this breeding program is not the improved populations A and B. but the hybrid seeds produced from random matings among A and B plants. The quality of the hybrid can be further improved with additional cycles of selection.

Let's say that we conducted three cycles of selection (six seasons). We would then have the improved populations A and B and, if satisfactory they can be maintained as such in isolation from each other. To make a population cross (to get hybrid seeds), pick a location where you plant both populations in alternating rows. rhen, as soon as flower buds form, rogue out all males from the population A rows. Allow the B males to randomly pollinate the A females and collect the seeds from the A females. This will be seed of the hybrid A/B. If you produce a large supply of hybrid seed and store it in a cool dry place, you can use this seed every season to produce your hybrid plants. Do not collect seed from the hybrid plants themselves since you will eventually observe a decline in hybrid vigor. Always go back to the original seed of the A/B cross.

Several people have asked what the hell turtles have to do with breeding. The answer, as we all know, is that all people have seven. I certainly have no monopoly on these creatures and feel compelled to inform the uninformed. Yes, we all have seven!

As once stated by J. McEno (a devout Poohist), Vilest we take ourselves too seriously."

First printed in Vol. 6, No. 4, page 42.


*********************************************
diagram of one cycle of reciprocal recurrent
selection:
*********************************************

Season #1 Each female is crossed with random pollen from Ao males and Bo females (on separate buds) to obtain topoross families and remnant half-sib families.


A B A B A B A B A B A B A B A B
| | | | | | | |
| | | | | | | |
TC1 TC2 TC3 TC4 TC5 TC6 TC7 TC8

...Eight separate topcross families from Ao females

---------------------------------------------

Season #2: Plant all 16 topsross families in a uniform environment with each family in a separate row.


A B A B A B A B A B A B A B A B*
| | | | | | | |
| | | | | | | |
TC1 TC2 TC3 TC4 TC5 TC6 TC7 TC8

*remnant half-sib family

...Eight separate topcross families from Bo females
---------------------------------------------

Evaluate each family for desirable traits.
Decide which females out of populations Ao and Bo produced the best topoross famines
(maybe three out of eight).
Reconstitute population A1 by mixing remnant seed from selected Ao females.
Reconstitute population B1 by mixing seed from selected Bo females. Plants from
populations A1 and B1 can now be crossed to make seed of the Improved hybrid
population. (TC in diagram means topcross)

******* back to the discussion **************

toadspitstout
Senior Member

What do you think of this hypothetical?
Using your example, a few growers decided to improve a strain they like using the method.

Each grower is given a lot of say 100 seeds, 60 to establish 5 moms the other 40 to be used for pollen.
The 60 are started and males are culled. The females are grown and selected based on the agreed criteria, ending with the five outstanding super moms.

Once the super moms are determined, the 40 other seeds are grown culling all females and allowing all males falling within the criteria to pollinate the super moms.

Seed kept in lots 1 thru 5. These are grown out to determine which lot is the best and within the agreed criteria. Again, five super moms are established from this best lot.

Each grower gives an other grower 40 seeds from their best lot. These are grown and females are culled and the males falling within the criteria to pollinate the super moms.

The process continues until the desired results are achieved or progress towards the desired goal ceases.

This would eliminate the need to collect and exchange pollen and would work with any number of growers.

Let me know
TS
______________________________

Hyb, get out the knight-hood equipment !!



quote:
--------------------------------------------------------------------------------
Originally posted by toadspitstout:

I was trying to keep the number of plants to a manageable number, I seem to be somewhat confused I guess by what you mean as "Best Male/ Female". In the hypothetical, I was anticipating growing the plants removing the males then letting the females finish. This giving a full growing cycle demonstrating phenotype, yield etc., then having the end results to test for taste and potency. This would require the taking of clones but this also would be valuable to demonstrate cloning ability of the strain. Also providing a bit of nice bud to keep me going through the project.
--------------------------------------------------------------------------------

You gotta ask yourself the question: "what is the likelyhood that i would select whimpy, weak plants to mother-up and save for future RRS breeding?" Also ask yourself "what is the likelyhood that if i inbred my 'top third' nicest growing males (combined pollen) with my 'top third' nicest growing females, that i would be losing valuable traits that only my whimpy, undesirable, cullable plants contained?" (breeding is heavily reliant on statistics, the study of probability.) Maybe these kinda questions can help you resolve the RRS strategy. You are most definitely a thinker when you are considering the side benefit of clonability in your strategy! Altho, in the end, with RRS, I doubt it would weigh very heavily as your ultimate F1 seedlot will have most or all individuals in a "keeper" genetic configuration. If you made 100,000 such seeds, what is the point of maintaining fragile mom plants for years? Properly dessicated and refrigerated, your super seeds would last for 20 or more years.




quote:
--------------------------------------------------------------------------------
Originally posted by toadspitstout:
The thought of using seeds not collected pollen; again was due to the number of plants and timing. In addition, pollen is more vulnerable to spoiling. At the point, the best females are determined, germination of the seeds for the males could be timed to grow out with the clones. The Female clone would be from separate plants not a number of clones of one.
--------------------------------------------------------------------------------

I already understood your clones each being unique, not all one type.
What are the chances that you will be excluding some important trait as you keep 5 clones out of 30 females (60 seeds planted, right?) as opposed to in-breeding small branches of each of the best 'top third' of your female population? (hint: 10 plants pollinated with mixed pollen from your 10 best males)
Pollen can be stored quite well if dessicated and refrigerated in a timely manner. Keeping mother plants going is alot more work than keeping pollen good for a buddy, or yourself, for a couple months. However, if you have a buddy to help out, you would want to both be on the same schedule.




quote:
--------------------------------------------------------------------------------
Originally posted by toadspitstout:
I am not sure I follow you here and the reason I reread the thread again. Yes the original seeds are from the same source and the first grow is just selecting the best from them, but from that point on the male side is coming from another grower. On the other hand, are you saying it would need an inclusion of other Females.
--------------------------------------------------------------------------------

You say that originally the seeds are coming from the same source, but then make them something seem different cuz something is coming from another grower. I know you well enuff, now, to know you wouldn't make a simple error like that. Maybe I'm mis-reading something.

This is crucial: What are the chances, during RRS, that breeding your healthiest males to your best females, continuing thru 4 or 5 cycles, that the quality of your males will go up? Now same question, but with quality of your males decreasing? Remember, your males are heavily influenced by the females. Male selection will become less of a problem as you realize they are continuing to arrive from genetically improved, inbred females, because the females have been selected by you! Beginning to gell for you?




quote:
--------------------------------------------------------------------------------
Originally posted by toadspitstout:
I understand this I just don't understand how you could determine the BEST without growing them out first. How do you do that?On the other hand is it that Family traits determine potency, yield. So selection of the individual is less important in these areas and you are determining Growth patterns mostly at this point enabling selection up to the point the pollen is released?
--------------------------------------------------------------------------------

In RRS, the second part of the cycle, the alternate growout, will be top-crosses only (the hybrids you made betwix the two populations). No seed will be made from this growout. Surely you will watch them grow and develop. Keep track of which inbred seedlot each individual came from. Then you will bio-assay! When it becomes apparent to you, based on your personal likes, which of the hybrids are just titular, GO BACK to the corresponding INBRED seedlots, combine equal amounts of only those seeds, and that will be the basis for the next round of 'top third' inbreeding and top-crosses (hybrid seeds). This just about covers the whole ball of wax.

As you choose to include certain inbred seedlots, and exclude others, you are limiting the amount of recessive junk that is kept within your inbred pools on both sides of the program. And at the same time your are ENHANCING the frequency of desirable alleles in both inbred populations (and the males!) that combine optimally for maximum hybrid vigor. In other words, the flavors and highs, the ripening times, growth characteristics and vigor will all be enhanced in the successive F1 hybrids. Side note: it would be interesting to see where you would end up by only breeding the 'worst third', etc would take you. I'm not about to mention BGS at this point I know, cheap shot! But I never claimed I was an expensive date.




quote:
--------------------------------------------------------------------------------
Originally posted by toadspitstout:
My goal in general is to grow the best I can with the room I have to do it. I am very excited about the possibility I might actually be able to improve the strains I have. My original plan would have lead me right down the "Inbred Dog" path, I was going to pollinate a clone of the best Female and a good Male.
--------------------------------------------------------------------------------

I am sure that Hyb and others are very proud of you right now! And yes, there is quite a good possibilty that you can improve your methods, might I even say RESCUE some of your older works? Pollinating with one male to your best female is only another founder event. Doesn't clean out any recessives and doesn't do much to enhance hybrid vigor in a population sense. Don't feel so bad about how you were once gonna do things (pay attention here Joey Weed), cuz everybody has to start out somewhere. Look at it this way -- at least the good genes are scattered thru your seedlots, preserved in that manner at the very least. Thank goodness you have not lost all your seeds in a fire or something.




quote:
--------------------------------------------------------------------------------
Originally posted by toadspitstout:
I have many strains that have similar lineage some very similar. Would it be better to combine these using RRS than to continue them along their lone path?
--------------------------------------------------------------------------------

Well, to get the most out of RRS, you need to start with the maximum variabilty. I'll give you a scenario and let you figure it out
Lets say you had a Cinderella line of seeds that was 100% homozygous at every allele, and you had another line of seeds that was Genius which was also 100% homozygous at every allele. Now lets say you wanted to do an RRS with them to get the perfect Apollo11 strain. Since all alleles are the same, for each population, your first top-cross (hybrid) is the most you can ever expect for the givens. There will only be one genetic outcome, and they will all be identical plants. End of program, no further improvement possible.
Now lets take some of your EARLY seedlots, and I can only speculate what you might call them. Maybe you can put your finger on 8 seedlots from the earlier years (before too much inbreeding and f3'ing happened) that you cherish. You can put 4 seedlots on each side of an RRS program and start saving the good stuff and throwing out the junk. If your wanna try for 10 females on each side, representing the 'best third', then you would need a total of 60 seeds planted, per side. You might see that one of your seedlots isn't doing that great, in the very first grow, and elect to include a female or two of them anyways, along with a male or two, just because you know there are things in there that you want, even tho the plants have plenty of other junk in them. It might possibly mean that you would ultimately need to do an extra RRS cycle somewhere down the road to help jettison the junk. But it would be worth it. I say that as long as you know and understand the principles of RRS, you can bend and tweek the method as you see fit. Steve in Switzerland may do it with 500 individuals whereas you might consider less!

When the RRS principles crystalize in your mind, you will realize that it is analogous to a 1000 piece puzzle where all the pieces are not static, but keep changing shape. When you METHODICALLY apply RRS, all your genetic puzzle pieces will OPTIMIZE into a plant that says "toadspitsout" !!

I'd like to say that this topic was originally introduced in the context of "how are we gonna save landraces?" It has evolved into "what can I do in my closet" which is okay, but doesn't do much for the Thai, Vietnam, and Cambodian etc landraces that have gone to krap. Those people also need access to this type of information so that they can bring it all back together as well. Seedvendors and certain book authors that frequent these lands are in a unique position to fill that need, as they communicate with growers over there. Even if it was a plastic laminated poster that described RRS in their native language, that would certainly be more direction than what they are currently getting thru whatever folklore they believe in.

"Odi profanum vulgus et arceo." -- Horace
_________________________________________

[Vic High]
Recurrent selection just means that a population is repeatedly exposed to the same selective pressure. Even the creation of the land races was through recurrent selection because each year the old farmers would take seeds from the plants that they felt were the closest to their ideals. In a simplistic sense, if you are inbreeding, you are more than likely performing a recurrent selection on the genepool. This is unless your ideas of what should be the ideal "look" changes from generation to generation, but for most seed makers, it wouldn't.

Now there are many breeding strategies that rely on recurrent selection to reach their goals, RRS is just one, cubing is another. Just keep in mind that these strategies are not a breeding program, they are nothing more than tools in the breeder's toolbox. Learning how and when to use these tools is what makes a sound breeder, not the tools themselves.

So back to the various forms of recurrent selection, which tools go with which breeding programs? Along with selective pressure, answering this question is one of the foundation of sound breeding. Joey and Toad, neither one of you have yet addressed this question with respect to your challanges. Once you do, you will understand why you are comparing apples to oranges and will be in a better position to contribute rather than challange.

As I hinted earlier, you must determine what your starting point is, and then determine where you want to end up. Do you want to create an IBL and the end product? Or do you want to create an F1 hybrid? Your reasons for producing the seeds will probably determine the answer to this question, if they are for personal backups, your primary goal is probably to just create IBLs. However, releasing IBL's to the seedmarket is like giving away your efforts, it's not hard to create a quality and consistent knockoff, therefore if your intent is to market the strain, then you just gave away your marketing edge. Therefore most intelligent quality seed breeders will stick to releasing F1 hybrids to the seedmarket. A breeder would have to invest a considerable amount of energy to recreate your work. And by capturing hybrid vigour, an F1 hybrid can be the highest quality product available to customer, better than any IBL.

A Sample Breeding Project - Creating a true F1 Space Queen

So lets say a breeder wants to create an F1 hybrid for the seed market, where to start? Well lets use a work in progress, my Space Queen as an example. I eventually want to create a quality and consistent F1 hybrid of Romulan and C99. I want my finished product to lean towards the sativa side of the spectrum that we witness in the test crosses I shared as Space Queen. I want the finished product to be uniform for the pine funk aroma of the romulan with it's trippy and pleasant high but I want C99's faster finishing time and the extra dimension it gives the high.

So now, what is my starting point? Well I have a special heterozygous clone (romulan) and a cubed (Bx3) seedline, Cinderella99 (C99). What is my first goal? To turn the two genepools into raw IBLs, the C99 is close, but the romulan is a long ways away.


Creating a Raw C99 IBL

With the C99, I have a genepool where each phenotypically important gene locus is homozygous in roughly half of the population and heterozygous in roughly half of the population. I would need to convert this so that my C99 genepool was completely homozygous for the phenotypically important traits that make C99, C99. I could grow out lots and find the individual male and females that are true breeding for all C99 indicative triats (this would require test crossing), but depending on the number traits (and more importantly, chromosomes) we are trying to stabilize, the number could get prohibitively high. Plus, we don't want to reduce our genetic diversity to just two plants so early in the breeding program. My preferred approach would be to grow out the C99s and created Bx3F2 families based upon half sib crosses (mixed pollen). Like say I grew out 100 C99s and selected the best 10 males and females that held the traits that I want from C99. It would be better to do controlled crosses between them, but then I would end up with 100 families to test out and I want to finish before they plant me in the grave. Maintaining genetic diversity is more important than perfect homozygosity, IMO. I have selected the best 20% of the population, that should be enough to ensure I haven't taken a step backwards with respect to potency and yield while still preserving the genetic diversity needed to pick out the more homozygous families. I would keep the seeds from each female seperate and grow them out so that I could determine which seedlings came from which parent. Roughly 20 females from each selected C99 seed parent should be enough to determine the female's relative homozygosity. Determine the best ... say ... 3-5, taking care that collectively, each important trait is relatively homozygous in at least one of the parents.

Now I would grow out more of the Bx3F2 seeds from the 3-5 selected seed parents as a group and repeat the process of choosing the best 10 males and females to create an Bx3F3 seedline. I could repeat the selection process and create a more truebreeding Bx3F4 seedline, but I think I will save further selections for later to preserve the genetic diversity I will need for the final stages of the breeding project. Now I have a raw C99 IBL that would be suitable for several F1 hybrid projects, not just the immediate one.

Creating a raw romulan IBL

Basically what I want to do is cross this clone with a male and then perform a recurrent selection process to isolate the romulan traits into a romulan IBL. To start, what should I use as a pollen source? Well to save time. I would first consider the romulan hybrids I've already made, romberry (romulan X blueberry) and space queen. I would give preference to romberry for 2 reasons. First, my goal is to preserve genetic diversity and I want to create as much hybrid vigour in my end product as possible, this is best accomplished by keeping C99 genes out of the romulan IBL. Another reason I would prefer to use blueberry as a pollen source is because blueberry's traits tend to dominate over romulan and are easy to distinguish from romulan's, therefore making them easy to select against. Individuals possessing romulan traits are more than likely homozygous for those romulan traits.

Now I have two avenues at my disposal, creating F2s, F3s, F4s, and F5s recurrently selecting for the romulan traits with each generation. If I was working in a field or large greenhouses and creating seedstock for either of those environments, this would be the preferred method. However, my seedline is intended for the indoor garden and the problem is that unless I manage high enough numbers with each generation, I risk a quick deteriation of the genepool's genetic diversity and the resulting inbreeding depression.

The second starting point to consider is the overly hyped "cubing of the clone" concept. Even though it's a tool meant to incorporate a new trait into an existing IBL, like disease resistance, I feel we could use the tool effectively here as well. What I like about it is that it allows the breeder to increase a seedline's homozygosity while maintaining the bulk of the genetic diversity found in the initial clone. The drop in genetic diversity between the P1 clone (romulan) and the cubed (Bx3) seedline is negligible.

So cubing will be the preferred starting point in creating the raw romulan IBL. But hey, before we dive in, what about the powdery mildew resistance I've witness with the original blueberry and some of the romberry hybrids created from it? The trait is just sitting there and a cubing is a perfect oportunity to incorporate this important disease resistance trait into the romulan IBL. In fact, since both romulan and C99 are susceptable to powdery mildew, I wouldn't be a responsible breeder of the Pacific Northwest if I didn't at least try

Now I've previously found that powdery mildew resistance is a polygeneic trait, but one that is highly heritable. Many of the genes involved in powdery mildew resisatance seem to be recessive. This means that I will need to add a recombination generation to each backcross generation to carry the powdery mildew resistance into the cubed (Bx3) population. If not, I will loose a minimum of 1/2 of the resistance with each generation, thus watering down that portion of the genepool. I've explained this in detail in the past, so I'll just give the rough outline. With each generation, the breeder would recurrently select for romulan genes and powdery mildew resistance. Potency and yield would be given consideration, but would not be the deciding factors since these genes will not be diminished by the cubing process. Their selection doesn't become highly important until after the cubing process is completed.

1st Cross: Romulan clone (P1) X Blueberry Pollen (P2)(pm resistance) => Romberry (F1)

2nd Cross: Romulan (P1) X Romberry Pollen (P2) => Bx1 (first romulan backcross)

3rd cross: Bx1 X Bx1 => Bx1F2

4th Cross: Romulan clone X Bx1F2 pollen => Bx2 (second romulan backcross)

4th Cross: Bx2 X Bx2 => Bx2F2

5th cross: Romulan clone X Bx2F2 => Bx3

Now we have a cubed romulan seedline that should grow out relatively uniformily and very similar to the romulan clone and with some degree of powdery mildew resistance. The next step would be to repeat the process used on the C99 cubed seedline to create the raw romulan IBL. Selecting amongst the Bx3F2 generation is a very important generation for fixing maximum PM resistance into the seedline, I would give it the most weight of all traits to select for and fine tune the romulan traits with the selections from the Bx3F3 generation. Therefore, I would probably take this population to the Bx3F4 generation before stopping but stop the C99 project at the Bx3F3 generation. These two raw IBL seedlines would become my starting points anytime I wanted to create an F1 hybrid with romulan or C99. Even if I wasn't interested in making or selling F1 seeds, but just wanted to create seed backup of a special clone for personal use, I would take the project to this stage. Finding another keeper clone from this genepool should be relatively easy.

RRS - fine tuning the raw IBLs for creating the F1 hybrid

Now there are a couple of approaches, one could simply cross the two IBLs and create somewhat of an F1 but the F1 would be somewhat variable. It would naturally carry some hybrid vigour, but this wouldn't be maximized because up to this point, selections have not been made for hybrid vigour. Afterall, creating an IBL inheritantly selects against hybrid vigour. One could reduce the variation of the F1 hybrid by furthering the inbreeding of the two raw IBLs independantly of each other, but still nothing has been done to select for hybrid vigour. Or for any other aspects of combining ability.

This is where RRS, Reciprocol Recurrent Selection, comes in, it allows the breeder to fine tune the raw IBLs AND select for combining ability AND hybrid vigour. All RRS means is that the breeder will base his selections of each generation of each IBL on how each individual combines with the other IBL. This is by far the best strategy to finish an F1 breeding program of any strategy discussed here. It is the only strategy the allows the breeder to select for combining ability AND hybrid vigour for a planned F1 hybrid, making the F1 hybrid realize it's full potential. Making the F1 competitive in the marketplace

Now there are several ways to carry out an RRS project, Hyb's suggested outline is just one. But for what it's worth, since I'm an indoor breeder, I would follow the mixed pollen strategy for the reasons I outlined in previous posts. When done, my final Space Queen product would be something like a romulan Bx3F6 X C99 Bx3F5. Assuming I used proper selective pressures, it would be a very uniform and highly competive F1 hybrid that would be tough for another breeder to create a quality knock-off.

I've left out several points, anyone is welcome to fill in the blanks.

[There is much useful info in my post but it also contains some fuzzy logic which reduces the effectiveness of the strategy. Where you able to identify my errors? If not, then you lack the ability to effectively use what I wrote. It won't be until you can identify my errors that the strategy I outlined would be of any value to your seed making efforts.]

_____________________________
[toadspitstout]
If you have not reached your goals then you continue with more cycles of RRS. The exceptional Individuals are the proof of the pool. The more frequently your are able to get the exceptional Individual shows how much more work is left. The goal in theory is every seed becoming an exceptional individual.
__________

FIN


I hope this turned out to be readable; doubtless a better job could have been done with the source material, but I saw a need for this information to be spread in a condensed form and I met it as best I could. If you would like to do a fancier, more coherent job of it I salute you for your efforts.

Good times and healthy plants to all!

-Telepod

[ButterflyDreams]

I have been following these discussions, but as you know many here prefer not to wade thru the muck.

From the scan I think you got all the good stuff.

There are several of these discussions ongoing on the boards.

While I do not subscribe to either of the extremes in the debate I can already see my personal philosophy has been altered.

Anyone interested and I'll put the links here.

So, now I want to hear our community.

What do you think? Don't be shy.

Smiles,
BD

[OzGrowa]

Great post but please dont do it again, hibe is a user at another system which retains ownership of that information (the same goes for Og too).

Ill leave it as yes, its very worthwhile, but any more posts like this will unfortunatley have to be deleted. If you can get hibe to post it here under his own name then that is fine, we have to be careful with these things as the system that hosts the message, owns it!

Regds,
OzGrowa

[ButterflyDreams]

My fault. I asked for it to be put here. I had sent a few to the links, but the politics and egos were too much to wade thru for many.

Sorry.

Once again my good intentions gets in the way of the admin stuff.

What do you think of the threat to the gene pool?

Smiles,
BD