Transform tetraploid into diploid?

[Ocimum]

Is there a way to transform a tetraploid into a diploid, for example by growing pollen?

I know that generally it is done the other way around, but diploids have some advantages as well.

[Joseph Lofthouse]

Culturing pollen is the easiest and most common way to half the ploidy of a variety.

[bill]

There may be other techniques as well.  What species are you working with?  For example, with potatoes, you can use a haploid extractor, which doesn't require any special skills other than patience.

The term for this process is haploidization, which will help when searching for information.

[Ocimum]

Thanks for the keyword, haploidization was the term I was looking for.

I will check out the book "Doubled Haploid Production in Crop Plants: A Manual" by Maluszynski et al.,

Best

[Walt]

In barley, and with some success wheat, pollinating with Hordeum bulbosum pollen can give haploids.  The F1 seed is a nomal hybrid,but as the cells divide, the bulbosum chromosomes are lost until none are left.  The seed head is a haploid of the non-bulbosum parent. 
Some other plants can be stimulated to make seeds by pollinating with pollen of a species that won't actually make hybrids with the seed parent.  The resulting seedlings are haploids.  This doesn't work with a lot of species, as far as is known.  But breeders don;t often check chromosome numbers on their failures, so it may be more common than is known.

[Walt]

Dr. George Liang of Kansas State U and his students found thatwhile triploid sorghum has very low fertility, the gametes it does make are haploid.  So is a triploid sorghum is pollinated by pollen from diploid sorghum, any resulting seedlings will be diploids.  You would have to grow several triploids and pollinate them to get even a seed or two.  But it can be done.

[bill]

In general, with a triploid, you are going to get a distribution of gametes ranging from x to 2x, concentrated around 1.5x, but not making up something as predictable as a normal distribution.  Depending on the species and what other sanity checks it performs against gamete configurations, you might find that only the euploid (x and 2x) gametes function in crosses with partners of equal ploidy, that some aneuploid (x+1 to 2x-1) gametes function, depending on which chromosomes they contain, or all of the aneuploid gametes work at least well enough for fertilization.  Whether or not the seedlings produced in aneuploid crosses will be viable is another matter, but they commonly survive in some species.

In species that have an EBN system (I don't know if sorghum does), most aneuploid gametes will be rejected because they don't have the required EBN chromosomes.  In species that lack this check, any extra chromosomes from aneuploids may still be dumped in the remarkable process of genome elimination.  So, either only euploid gametes make it to fertilization or only euploid zygotes come out of fertilization.  In other cases, aneuploids make it all the way through and manage to produce a viable progeny, although it seems that the extra chromosomes most often get lost in the next generation if the plant is fertile.  In short, nature finds a way.

My expectation would be that you would get haploids reliably from triploid x diploid crosses when the 2x gametes produced by the triploid are fully incompatible with the x gametes from the diploid for whatever reason and perhaps the aneuploid gametes are all resolved by genome elimination.

I really love this stuff.  It is neat to see how nature makes rules and then immediately sets about finding ways to break them.