Direction of parenatge and mature plant size

A question was posed on Rose Talk whether the direction of a cross had any significant effect on the mature size of the plant. To illustrate what I mean consider the fictitious cross of ‘Mr Lincoln’ with ‘Rise n Shine’ (I chose these two deliberately because ‘Mr L’ is such a tall rose and ‘Rise n Shine’ is a miniature). Would the resulting seedlings be taller (on average) if I chose ‘Mr L’ as the seed parent or as the pollen parent? My reply (see link) was that I had never heard of any differences in mature size caused by varying the direction of the cross and that it would probably depend what the parents were and that different roses behave in different ways depending on their heritage. I also said I would ask here to see if anyone had noticed this. I can’t say that I’ve ever noticed any differences.


Thought I’d mention that I don’t mean size due to miniaturisiation in case anyone was wondering.

That’s a great question. In the Spring 2002 newsletter I wrote an article summarizing some of the possible reasons why direction of a cross may influence the segregation of traits. For the most part I agree there shouldn’t be a strong influence. Modern roses have limited cytoplasms I suspect. So when we are talking about Rise 'N Shine it should have the cytoplasm of Little Darling. Cytoplasm has mitochondria and chloroplasts of the female parent with few exceptions and they contain some of their own DNA. The interaction of that with the nuclear DNA can lead to some traits being expressed a little differently depending on who was the female and contributed the cytoplasm. There are some viruses that appear to be seed transmitted too and that can play a role on performance probably being more common through the female side I suspect. Meiosis can be variable too and pollen competition influencing things. For instance, if one uses Knock Out as a parent which is triploid, more gametes that may be a bit weaker can be effective in fertilization when on the female side, than if on the male side where they may be outcompeted, especially if they are aneuploid.

Just some thoughts. The Spring 2002 article summarizes some of these and other points in more detail.


Agree with David that there shouldn’t be much difference, but what we have often found in the crosses we have made is that when we use the larger plant (HT, F, LCL, etc) as the seed parent and a mini as pollen we find a larger percentage of the seedlings to be of a miniflora size. Reversing that cross has often yielded for us a greater percentage of minis.

Will have to use the super RHA newsletter CD and go back and read his article again. If you don’t have one of these, contact Larry Peterson.


Yes… I’ll have to try and get a copy of the CD now!

If you were to design an experiment to test this theory… how would you go about it? I was thinking you would have to use fairly well known/understood diploids to reduce the total number of genes you were playing with and then perform the cross both ways and grow the seedlings side-by-side for comparison. Do you think it would be a variable thing and that some roses, or groups of rose, may show different results? For example if a species, or near-species rose was used or would rugosa have a different effect than gallica that is different to wichurana etc? Do you think ploidy might exacerbate the effect?

I decided to sleep on this one last night. The only plausible notion I could come up with that might allow for different sized plants based on parentage direction is that maternal or paternal imprinting (methylation of DNA affecting gene expression, not permanent changes in the gene sequence) might alter gene expression. Why it would do so I have no clue.

I wrote an article on genome bias in hybrids last quarter. Genome bias is teh preferential expression of genes from one parent over those of the other. A size difference or other obvious phenotype variation, like thorniness or leaflet number, could be produced by genome bias. The cause of that is unknown. It may ahve to do with linkage disequilibrium, so it may be true in a cross done either direction. It seems tha whole pathways or types of metabolism work as a unit which is controlled by whichever parent happen to be dominant. But for cotton, the species well studied and that I discussed, there is only one hybrid plant. It is a very infertile cross. so we don’t know if this is a general thing and if it works both ways.

I assume that if there is some size bias or other inequity of gene expression, it will be different for all different combinations of species. So for instance the idea that spinosissimas show up dominating things like thorniness and leaflet multiplicity, could be genome bias. On the other hand it might be paternal imprinting (DNA methylation) if we only look at spins as pollen donors. We have to worry about breeder bias here.