Slight tangent, but less scrupulous individuals might access that paper on Sci-Hub - the infamous research publication piracy site described in this article which somehow continues to evade being shut down.
Sci-Hub had failed me with other very old studies, so I hadn't tried it with this one. Turns out, access was obtained.
And, having read it... oh lord. Wow, really?
"She died. She had eaten rhubarb. I heard rhubarb is toxic because of OA. Therefore, she died because of OA poisoning." This is pretty much it. No causal link between OA and death was determined.
I'm going firmly into "the link between rhubarb leaf's toxicity and OA is completely unfounded" position. It's just hearsay quoting hearsay quoting hearsay. The authors don't even know how much OA is in there. Heck, the article IS WRITTEN AS A QUESTION. Here's a direct quote: "What the content [of OA] in the leaves is, I do not know." He basically just states "Hey, could this be OA poisoning?", and the editor replies "OA can be toxic." That's all.
It's a bit of a struggle not to derail this thread, but I'll succumb to play devils advocate - hopefully - succinctly. I wonder how successfully one could make a case that truly open source plant breeding needs to be boot-strapped from the "ground up" (pun intended) starting with open-access/permissively-licensed research. Conversely, could an open source variety actually be considered encumbered if the knowledge or techniques that lead to its creation are themselves encumbered by intellectual property? How about if the process of creating it involved intellectual property violations that themselves aren't present in the resulting variety?
Presumably, there would be no issue if someone learned about working Sci-Hub links being easy to find with Google through a non-endorsing post on this forum and subsequently used that knowledge to learn about the OA toxicity which eventually effected the direction of their open source planting breeding efforts.
I don't know - IANAL - but am curious where the line would be drawn - probably a question for a different thread though... 
Scientific articles aren't an IP issue. Besides, that article is so old, even if it was treated as a book, IP would probably be extinct by now. But that's not the point, the knowledge in said articles isn't intellectual property. These aren't patents. You don't owe anyone anything for making use of something published in a scientific article.
And patents are not major obstacles to breeding. I know of one patent regarding breeding techniques, and there are probably a few others, but these are super rare and not applicable to almost every breeding scenario (the example I have in mind is the use of certain chemicals to cause sex reversal in hops).
Even plant variety protection, for the countries where it was granted, don't restrict the use of a genotype for breeding in any significant way).
The only significant patent obstacle to breeding is an utility patent, which to the best of my knowledge, is an American aberration, and has no force of law outside of the United States. In the grand scheme of things, even these are relatively rare
And all patents end up expiring anyways.
Back to the topic of Rubarb breeding.
The remaining question for me that remains unanswered is whether the properties that make 'wild' (ie. non cultivated for food) rhubarb varieties (other Rheum spp.) is the same as that present in Rhubarb cultivated for food (ie. Rheum x rhabarbarum).
I do not think it can be assumed that Oxalic acid is the only factor involved, or is present in a similar amount. There may be other factors involved, and it appears that the available literature does not adequately answer questions about this. 'Safety' or 'edibility' for such things may be similar to glycoalcaloid content in potatoes. If only a small amount is eaten, perhaps it is 'safe' because the human body can process out the toxins without a healthy person becoming sick. But eaten in higher amounts, elevated glycoalcaloids can make a person sick. Generally for potatoes, for example, there are safety thresholds to this, and as long as the total accumulated amount is not ingested, you will not get sick.
I would not be confident in working with any of the other Rheum species without information being available that does not appear to be available. Maybe other people would be, however. It poses interesting questions. I probably will stick primarily to Rheum x rhabarbarum accessions and let others deal with breeding away from laxative properties, whatever the reason or chemicals involved.
The safer assumption is that OA is *not* the only factor involved. Other species, not as well known as the already poorly understood cultivated rhubarb, could have:
1) Different concentrations of rhubarb's toxic chemicals
2) Different distributions of rhubarb's toxic chemicals
3) Different chemical compositions including absence of rhubarb's toxic chemicals or presence of new toxic chemicals
Rheum rhaponticum is the only other species I have found that is reported to have its petioles consumed, so if I were to attempt an intraspecific cross, I think that's the only species I'd be willing to try with. And even then, I'd do a bit more reading before proceeding. Otherwise, it would definitely be safer to only work with Rheum rhubarbarum, even if that narrows the gene pool significantly. On top of which every accession is supposedly tetraploid, while it'd be nice to have access to diploid varieties, and I doubt someone worked out a protocol for anther culture of rhubarb (though I didn't look yet).
