Two seedlings from one seed

Twice this year I’ve had two seedlings come out of one seed. I checked very carefully because I couldn’t believe it at first. In both cases, one seedling was normal sized, and the other was much smaller. Has anyone else seen this? What could cause this?

That’s great Jim!

I’ve been saving twins the past couple years and am axious to confirm their chromosome number when I have a chance. Sometimes a secondary embryo forms from a synergid (a cell with the haploid chromosome makeup like the egg) and will be haploid. Sometimes there can be two ovules within one seed coat though too. If both the embryos were in the same testa (papery covering) chances are high that the smaller one is a haploid. If they each had their own testa the chances are high for two separate ovules and them not being haploid. Only testing for sure will tell though what chromosome number each plant has. I’ve generally found twins with their own individual testas, but I have found a couple where both embryos were in the same testa. Do you remember if they had their own separate testas, or if they were together in one?



Wow, if the smaller ones are half the chromosome number of a tetraploid hybrid tea parent, you have very important seedlings. They can be used to introduce hybrid tea genes into species roses and get fertile offspring.

Good luck.


This is really exciting! I had one twin seed a few years back and actually pulled the weaker of the two out, as I was afraid to try and seperate them and thought it was detrimental to have them growing so close together. It was only later that I learned from David what the situation might be. Finding a haploid from a tetraploid rose would be as useful for breeding exotic diploid hybrids as using doubled diploids on tetraploids. It opens up a whole new world of possibilities!

Who are the proud parents?


The most recent twins definitely shared the same testa. I’m not sure about the previous pair. I was hoping the smaller seedling might be diploid, and I’m thrilled to hear it could be!

The most recent pair was from the cross Precious Platinum X Veteran’s Honor. The other came from a Tournament of Roses seed. I don’t remember the pollen parent off the top of my head.

David, after hearing Randy’s experience, I suggest that one of the 3 of you (Jim, Randy, or you) (or all three in a combined article)write an article for the newsletter about what 2 seedlings from one seed may indicate as probably many people (including me) have done just what Randy did. I did not realize that it must be relatively common since all three of you have experienced it.

Ditto to Henry. This is something way to cool for me to never have heard about before, and you ought to get the word out.


Well, I’m sure not the one to write anything on the topic. In fact, I still don’t really know how this can happen, other than a nonreproductice portion of the egg must somehow get “tricked” into behaving as if it were. I still don’t even know what a testa is. It sounds too much like a body part you try to protect when playing baseball. Ha.


Randy, the reason that I suggested the three of you is that the fact that 3 people have observed it is an important part of the message. I knew it could happen (not that it would appear as a twin, but that an unfertilized seed could occur) but assumed that it was so rare that I did not even consider it.

The testa is the inner papery seed coat. Never heard about there ever being two seedlings inside one though.


After opening at least 2000 seeds last winter, I saw about 5+ with double embryos. I can’t remember if any were in the same testa, at least some were not. I didn’t think much of it, since the rose “seed” is actually a one-seeded fruit (achene), with the embryo and the testa the whole seed, and the woody outer layer (pericarp) the fruit wall. Since most plants produce multiple seeds per fruit, it isn’t surprising to see the occasional double-seeded fruit in roses. I didn’t keep track of which ones were the doubles, unfortunately, and I killed an awful lot of seedlings. After hearing about this whole diploid seed issue, I am watching my seedlings carefully for anything out of the ordinary!

It’s great that so many people have observed twin seedlings. It’s probably more common than we realize. One of the major challenges I think is observing twins while they are still young enough to know that they came from the same seed. I germinate seedlings in plastic baggies and transplant them while the seedlings are still very young. By germinating seeds this way I can more easily find twins and “baby” along the weaker twins. It has been common to find twins within the same testa that are so small that they don’t survive (about 1/16"). They appear kind of translucent. Finding relatively larger twins that are more prone to survive seems rarer. I’ve noticed translucent smaller embryos in seeds from diploid rugosas and polyanthas even. So far I was only able to get smaller twins from the same testa to survive from tetraploid parents. Perhaps the smaller embryos from the diploid parents are monoploid and it is very difficult to obtain some that would survive without any deleterious recessives that are not masked. The smaller twins from the tetraploids may be diploid and dominance can mask some of the deleterious recessives.

I wrote an article in the RHA newsletter in 1999 on haploids (their use and extraction) and highlighted different means of extracting them including twin embryos. It’s been a way that corn and flax haploids have been extracted. Through learning this technique in a cytogenetics class a few years ago I have kept an eye out for twins in roses. I and many others have found twin embryos in roses. In addition, unfertilized eggs, but a fertilized central cell (endosperm develops from here) is another way. By crossing a repeat blooming tetraploid rose with a rose that is homozygous for one time bloom one should expect true hybrids to be one time blooming. If any seedlings are repeat blooming one can suspect the plant developed from an unfertilized egg and may be haploid. Of course one needs to confirm the chromosome number to tell for sure.



pollen size can be directly related to chromasome number, or more correctly ploidy level, as this is generalization is only accurate to the nearest full set of chromosomes, within a specific plant group or species. In any case Looking at the pollen of these plants under a microscope should be an easy way to determin Their relative chromosome complement.

Pollen of a haploid plant should be half the size of that from a diploid plant, which would have pollen half the size of a tetraploid plant. This should work well in this case because the plants are directly related. The only problem would be if there are meiotic irregularities in which case a plant might produce multiple types and therefore sizes of pollen with varying chromosome compliments. I am not sure if this is the case with roses though it is often true of plants that are derived from multi-species origin especially first generation hybrids.