In message , Nick Maclaren
writes

In article ,
Stewart Robert Hinsley writes:
|
| There's not much botanical available online, but I found a 2007 paper
| with assorted chromosome counts for subgenus Dracunculus. This reports 4
| cytotypes of A. dracunculus (diploid, tetraploid, hexaploid and
| decaploid), and gives A. dracunculoides as hexaploid. Wikipedia says
| that French tarragon can't be propagated by seed, but elsewhere I see
| claims that it's just a matter of poor seed set in cool climates.

Boggle. I didn't know about the chromosome counts. That sounds
bizarre even by plant standards.

There's many plant "species" with multiple cytotypes, and botanists
historically have not commonly recognised cytotypes as species. (It
seems to me that this attitude is beginning to change.)

For example the Lesser Celandine has (at least) three cytotypes -
diploids (ssp. ficaria), tetraploids (ssp. bulbifera) and triploids.
Contrary to Stace, I reckon this to be a clear case of two species - he
says that triploids are completely sterile. (Compared to a couple of
triploid mallow hybrids [sylvestris x durieui and alcea x moschata] I
grow, which have seed set at about 0.5%.)

The Yellow Archangel and Rosebay Willow Herb have diploid and tetraploid
cytotypes. So does Hibiscus trionum. And then there's the various
tetraploid cultivars, such as Myosotis 'Azur'. Some sterile Hibiscus
syriacus cultivars were created by forming tetraploids with colchicine
treatment, and backcrossing to the diploids.

In the Centaurea jacea complex of knapweeds (C. jacea and C. nigra) the
species boundary appears to lie not between the jacea- and nigra-morphs,
but between the diploid and tetraploid cytotypes.

The existence of multiple cytotypes suggests the existence of multiple
species, but is not conclusive. I don't know of an unambiguous case of a
plant species polymorphic for ploidy, but I wouldn't be surprised if
there was one. There are permanent odd polyploids which segregate m
haploid genomes into the embryo sac, and n haploid genomes into the
pollen - the best known case is the dog roses, but there's maybe half a
dozen other cases known. One could imagine a plant in which the same
segregation happened in triploids, except not specific to either sex, so
diploids, triploids and tetraploids were freely interbreeding.
Alternately if triploids produced unreduced gametes, and tetraploids and
hexaploids had regular meiosis, you could have a mix of diploids,
triploid, tetraploids and hexaploids - and possibly more depending on
what pentaploids did.

| Botanists don't seem to agree as to whether there is one species or two.

Or four, or .... Until and unless there is some evidence on whether
the various ploidies are inter-fertile, that sounds like angels on a
pinhead material.

| There is a paper on Artemisia phylogeny, but it's behind a pay-wall, and
| I doubt that is sheds light on the dracunculus/dracuncoloides issue.

Yeah, me too :-(


Regards,
Nick Maclaren.

--
Stewart Robert Hinsley