History is really OLD

A few days ago I was going through one of my “To Do” folders and came across this article, Is Ancient DNA Research Revealing New Truths — or Falling Into Old Traps? (2019) by Gideon Lewis-Kraus One passage in particular caught my attention:

About 5,000 years ago, a “relatively sudden” mass migration of nomadic herders from the east — the steppes of eastern Ukraine and southern Russia — swept in and almost entirely replaced the continent’s existing communities of hunter-gatherers and early farmers. These newcomers were known to exploit many of the cutting-edge technologies of the time: the domestication of horses, the wheel and, perhaps most salient, axes and spearheads of copper. (Their corpses sometimes featured cutting-edge wounds.)

This reminded me of another westward migration that occurred at around the same time.

The earliest historical records on Mesopotamian cuneiform tablets indicate that rose became known to humans about 5,000 years ago. A clay tablet about Sargon I, King of Akkadia (2684-2630 BCE), records that the king brought rose saplings during his military campaign to the countries across the Tigris River. Because he formerly lived in the ancient city of Ur near Babylon, his trip was most probably to southeastern Anatolia (present-day Turkey).

It’s been a long time since my High School Western Civ class, but I seem to recall that it all began in Greece. However, I don’t recall any reminder that the earliest “Greek” philosophers/scientists hailed from Miletus in Asia Minor (=Anatolia). E.g., Thales of Miletus, (d. 548–545 BCE). Ancient Miletus is now part of Istanbul (not Constantinople).

This historical trivia is rose-relevant because Pliny the Elder [AD 23/24-79] mentioned, “… the rose of Miletus, the flower of which is an extremely brilliant red, and has never more than a dozen petals.”

This brief note is not enough for a positive identification, but it does agree with the Velvet Rose. Of course, I cannot insist that the Holosericeas, and the possibly related Tuscans, were included in Sargon’s rose plantations. But as Linnaeus wrote in his Hortus Cliffortianus (1737):

We are a little worried about the catalog of the Roses, so long as we do not know how to distinguish the species; nor are the species determined in roses by anyone, nor can they be described by anyone other than the one who examines the birthplaces."
That said, it seems clear enough that > Rosa gallica > and > R. Damascena,incarnata,provincialis > are not natives of France, though they may have become naturalized there.

In addition, its lack of genetic diversity suggests that Rosa foetida is not native to Persia … or anywhere else. The most genetic diversity, such as it is, is now found in Turkey.

As for displaced species, here’s an interesting tidbit:

Ann Bot. 2015 Feb; 115(2): 275–291.
Phylogeny and biogeography of wild roses with specific attention to polyploids
Marie Fougère-Danezan, Simon Joly, Anne Bruneau, Xin-Fen Gao, and Li-Bing Zhang

The only sub-Saharan species of the genus Rosa, > R. abyssinica> , is resolved by chloroplast data as embedded in the Cinnamomeae clade but clearly shows the synstyly consistent with R. sect. Synstylae. The nuclear GAPDH sequences resolve this species as closely related to > R. gallica> (R. sect. Rosa) and European species of R. sect. Synstylae, suggesting a hybrid origin.

So, a Musk rose, of a sort, had a Cinnamomeae grandma.

And while I’m here, I should mention that cotton has been cultivated in the Americas (all three) for more than 6,000 years. The fun fact is that the various species are all allotetraploids, and share one common parent: Gossypium herbaceum, a species native to Africa. Somebody was doing some traveling, back then. By the way, cotton was grown in Sudan (next door to Abyssinia/Ethiopia) long before the Egyptians caught on.

Thank you for sharing Henry Kuska!
That is really fascinating, both about the roses and about the cotton!
Rosa Abyssinica sounds like a really interesting one, I had heard of her but didn’t realize she could repeat.
I’m going to have to hunt around and see if I can find a seller in Europe. Considering the increasingly hot and rainless climate of Malta, she just might like it here.

Continuing onward, I learn that the chloroplasts of Rosa gallica and R. arvensis (along with R. abyssinica) are nested within the Caninae species. Seems like an odd place to be nested.

But as I’ve mentioned before, Heslop-Harrison (1954) found that the odd Canina-meiosis could break down when crossed with a “regular” species.

Professor Heslop Harrison emphasized that the F1 lots, whilst conforming, in a general sort of way, cytologically to the usual Caninae pattern, in their later meiotic stages on the female side showed important anomalies. As a result, amongst the seedlings, orthoploid plants were secured carrying chromosome complements of 14, 28, 35 and 42. Thus it was clear that a new polyploid series had been evolved by a distinctly novel mechanism.


All those offspring would carry the same chloroplasts, of course.

This is not proof that Rosa gallica and R. arvensis received their chloroplasts directly from Rosa tomentella as Wissemann, et al. (2005) suggest, but I’m not going to speculate about Gallica and Arvensis giving rise to the whole section of Caninae section.

Furthermore, because the F1 hybrids, as Heslop-Harrison reported, conform in a general way to the usual Caninae, I will speculate. Suppose the F1 hybrids are pollinated by the original maternal species. The foreign chromosomes are “in there”, and the original pollen genome is brought back to re-stabilize the normal Caninae-meiosis. A new species?

In this case, it may not be necessary to assume that the Caninae-meiosis evolved twice: once for the species with fragrant leaves, and again for the non-fragrant types.

I was re-reading one of Heslop-Harrison’s earlier papers, and was struck by a footnote.

  • Although to my mind this does not explain the species-type, it may nevertheless be the correct view. Independently of these possible linked genes others do exist in Rosa. That a detailed analysis of their genetical behaviour has not been made depends on the slow growth of the plants; at any rate the attempt is being made. The following characters appear to be linked: glandular leaves are always biserrate; forms with deep red flowers have hairy foliage; erect and suberect, persistent and subpersistent sepals always accompany woolly flat heads of stigmas and early ripening fruit, whilst dilated and broad bracts are correlated with short peduncles.

I’m not sure whether he meant that these apparent linkages are found only in the Caninae species, or in Rosa species in general. For example, Roxburgh’s Rosa glandulosa is commonly lumped in with R. moschata Herrm. even though the leaves were biserrate rather than simply serrate. Would they also be fragrant? Would such a species augment the odor of, say, R. cinnamomea which is supposed to have cinnamon scented leaves?

Finally, Kohnen et al. (2009) suggest that these fragrant glandular trichomes play a role in resistance to rose rust.

Yes, thanks Henry :wink:

How do you dredge up these papers, are you near a libary with an FPS subscription?

it may not be necessary to assume that the Caninae-meiosis evolved twice

If, indeed, Caninae-meiosis did evolve twice then would comparing genomes between ITS types help reveal how such squirrely meiosis arises?

I wonder if there are hybrids between the two ITS types. Is there enough data at HMF to divide the Caninaes into those buckets?

BTW, Latin is Greek to me but would it not be ‘Sericas’ rather than ‘Sericeas’?

It’s this new-fangled thing the kids are calling “inter-tubes”, or some such.

There are some really terrific libraries on-line with great collections scanned and OCR’ed for our convenience. Some are are easier to access than others, of course. I have a list of some I use on the right-hand side of my What’s New page.

https://booksc.org/ is an especially good source for papers, rather than whole journals. I use it frequently.

Then there is the Google library that lets me search for on key words. This is wonderful fun until you run into something like ‘Gloire des Rosomanes’, which has way too many alternative spellings for such a simple rose.

I do have a tendency to fall down rabbit holes. There was the time Tom Silvers mentioned “Terra preta”, which is a dark, anthropogenic soil that’s a big deal in Brazil. I took a quick look, then browsed further, and eventually found myself exhuming records going back to 1603, in Wales.

As for Sericas/Sericeas, you are probably correct. My eyes aren’t what they used to be, and my proof reading has suffered.

It is odd, to say the least. The nearest analogy I’ve found is in wheat.

First, we commonly regard bread wheat (Triticum vulgare) as a hexaploid. But when an ovum is induced to develop without fertilization, the chromosomes do not pair off. They’re just 21 univalents hanging around unattached.

That is the less-weird part of the Caninae weirdness. A “gene” is present, somewhere in the genome, that inhibits pairing enough to prevent the homeologous chromosomes from pairing. Knock out such a gene, and the chromosomes get tangled in pairs and rings.
Riley (1958), Riley & Chapman (1958)

I’m guessing there must be more than one such “gene”. I guess further that there are likely some such “genes” that enhance pairing, at least in come cases. Chromosomes in octoplod roses must pair off, but (as I recall) form only one crossover per pair. Temperature seems to influence pairing frequency, so plants in the arctic region may need a little boost … but not so much as to let homeologous chromosomes get together.

Jiang (1998) describes the case of a chromosome that passes to 97% of pollen grains, but only 20% of ova. “We suggest that the long arms of homoeologous group 5 chromosomes in Triticeae species carry genes that affect their transmission through male gametes.”
I assume that this chromosome would perform “normally” when paired with itself.

I’m way far out over the edge here but I’d bet that, rather than genes being involved, there are fundamental physical properties at work involving electrostatics. The net effect would be similar where sequences govern function but the effect would be directly on the binding energies between and among the packaged chromosomes. I’d be surprised if there isn’t already some body of knowledge in this regard.

I’m right there with you. I had a brief article that I thought explained what we need. I read it again, and what I want is not there. I got more of the article, added it, and still don’t know any more than I did. That’s frustrating. The bit was from Darlington and Mather. I feels like they were daring their readers to find anything that might prove useful. Like a dream where I reach for a beautiful egg-sac and end up with a handful of Ascarid worms.

They did distinguish the two forms of cell division, and intimated that the formation of blood cells (white cells and red corpuscles) in different tissues is directly analogous to male and female meiosis … but somehow manage to spill much ink without clearing up much in the process. Read them at your own risk.

I did upload some parts of another article that show more examples of how symmetrical and asymmetrical cell divisions produce different cell types.

Then I returned to another paper that is worth noting:
BioScience 62(7): 623-624 (July 2012)
**Meiosis Is Not Gender Neutral **
Root Gorelick

Even more insidious, in all animals and plants, female meiosis is asymmetrical, whereas male meiosis is symmetrical. The symmetry of male meiosis seems to be universal in eukaryotes, with all four meiotic products having the potential to be passed on to the next generation, as is depicted in all commonly used first- and second-year university textbooks. Although on average, most sperm have virtually no chance of having their chromosomes passed on to the next generation, all four products of male meiosis seem to be equally viable.

By contrast, only one of the products of female meiosis goes on to form a functional egg cell (which, in plants, may be after several intervening mitotic divisions) and has the opportunity to pass its DNA on to the next generation.

One problem I’m having is understanding what different authors mean by symmetrical/asymmetrical. Sometimes the words refer to cell division, but in others it seems that the authors are referring to outcomes. That is, in male meiosis, all for haploid cells seem to be equally viable, whereas in female meiosis they take on different roles: ovum, polar cell, cell that fuses with polar cell.

That’s about all I can wrap my head around this morning. Although I was annoyed to find a 21st century author mentioning Mendel’s “law”.

And I had another look at Koopman (2008). Oh my head! Why should I care about guesses made by Wylie (1954)? She borrowed older guesses made by people who, generally speaking, knew no more than she did.

According to Koopman et al (2008), ‘Blush Damask’ and ‘Superb Tuscan’ are closest cousins. Likewise, ‘Communis’ (Centifolia?) and R. alba suaveolens. I happily accept that one, because from what I’ve read the original Centifolia of the Nabateans and of Pliny the Elder was none other than a very double form of our Rosa alba.

But what about all those Rosa gallica bushes allegedly growing wild in Europe? Are they ancestors of any garden roses? Or are they derivatives of Rosa tomentella and, maybe, the old Holosericea.

Here is one of the supplemental files from Koopman that gives the actual cultivars used in the study in case you want to speculate more deeply on relationships.

I had another look at Koopman

As for Wylie, well the author list of that paper is a 2008 Who’s Who of rose genetics so apparently they had a consensus on whose evolutionary tree they were raining on.

I wonder if any other plants are known to have similar meiosis to the caninae. I wonder, too, what the backstory of the discovery of canina meiosis is. Why did it come to Hurst’s attention when it did. Like, why was he studying roses and not, say, hibiscus or rhododendrons? Has similar meiosis been identified in other plant species?
appendixS1.pdf (29.8 KB)

I am not aware of any other genera with Caninae-like meiosis. I have reports of hybrids with a chromosome refuses to behave as the neo-Mendelists would like, by ending up in pollen grains FAR more often than it does in ova.

As for Hurst, maybe he just liked Roses. He had established his credentials as an Orchid man, contradicting various assumptions about heredity. He had a large paper published in 1900, just as Mendel was getting a bit of notoriety. He found that the two parents did not contribute equally to the hybrid offspring. That’s messy, but closer to truth than the idealized Mendelian model. Hurst wrote:
“When several hybrids from the same pair of species are compared together, this variation of the parts, or ‘Partial Prepotency,’ as I propose to call it, becomes even more apparent and more diverse.”

By the time I learned of this, I had already been calling the phenomenon “Elective Expression”, as a nod to Dr. Prosper Lucas, who wrote the book (literally) on Natural Heredity.

Oh, and Hurst grew up in the nursery business. At one time he was the “Son” in Hurst and Son.
PS: Hurst also studied heredity in horses, rabbits, Oenothera, and so much else. He was living at a time when chromosomes were being counted and studied, so of course he was all over them.

BTW, it was O. Winge (1917) who raised the issue of sterile diploid hybrids that became fertile, and oddly true-breeding, when the chromosomes doubled. Being a careful observer, accustomed to examining lots of characteristics, he could see traits of two distinct species superimposed in tetraploid species. Not all the traits were expressed, of course, but the groups were apparent.

Other people of the time felt uncomfortable discussing even a few traits that seemed to be correlated. That violated Mendel’s “Law” of independent assortment. It is worth noting that Van Mons was discussing such correlated characters a century earlier as he revolutionized Pears. Academic scientists were way behind the curve on this matter.